Human Remains from the -Holocene Transition of Southwest Suggest a Complex Evolutionary History for East Asians

Darren Curnoe1*, Ji Xueping2,3*, Andy I. R. Herries4, Bai Kanning5, Paul S. C. Tac¸on6, Bao Zhende7, David Fink8, Zhu Yunsheng5, John Hellstrom9, Luo Yun7, Gerasimos Cassis1, Su Bing10, Stephen Wroe1, Hong Shi10, William C. H. Parr1, Huang Shengmin11, Natalie Rogers1 1 School of Biological, Earth and Environmental Sciences, University of New South Wales, Sydney, New South Wales, , 2 Yunnan Institute of Cultural Relics and Archeology, Kunming, Yunnan, China, 3 Archeology Research Center, Yunnan University, Kunming, Yunnan, China, 4 Archaeomagnetism Laboratory, Program, School of Historical and European Studies, La Trobe University, Melbourne, Victoria, Australia, 5 Honghe Prefectural Institute of Cultural Relics, Mengzi, Yunnan, China, 6 Place, Evolution and Rock Art Heritage Unit, School of Humanities, Gold Coast Campus, Griffith University, Southport, Queensland, Australia, 7 Mengzi Institute of Cultural Relics, Mengzi, Yunnan, China, 8 Institute for Environmental Research, Australian Nuclear Science and Technology Organisation, Sydney, Australia, 9 School of Earth Sciences, University of Melbourne, Melbourne, Victoria, Australia, 10 State Key Laboratory of Genetic Resources and Evolution, Kunming Institute of Zoology and Kunming Primate Research Centre, Chinese Academy of Sciences, Kunming, China, 11 Youjiang Nationalities Museum, Baise, Guangxi, China

Abstract

Background: Later Pleistocene human evolution in East Asia remains poorly understood owing to a scarcity of well described, reliably classified and accurately dated fossils. Southwest China has been identified from genetic research as a hotspot of human diversity, containing ancient mtDNA and Y-DNA lineages, and has yielded a number of human remains thought to derive from Pleistocene deposits. We have prepared, reconstructed, described and dated a new partial skull from a consolidated sediment block collected in 1979 from the site of Longlin (Guangxi Province). We also undertook new excavations at Maludong (Yunnan Province) to clarify the stratigraphy and dating of a large sample of mostly undescribed human remains from the site.

Methodology/Principal Findings: We undertook a detailed comparison of cranial, including a virtual endocast for the Maludong calotte, mandibular and dental remains from these two localities. Both samples probably derive from the same population, exhibiting an unusual mixture of modern human traits, characters probably plesiomorphic for later Homo, and some unusual features. We dated charcoal with AMS radiocarbon dating and with the Uranium-series technique and the results show both samples to be from the Pleistocene-Holocene transition: ,14.3-11.5 ka.

Conclusions/Significance: Our analysis suggests two plausible explanations for the morphology sampled at Longlin Cave and Maludong. First, it may represent a late-surviving archaic population, perhaps paralleling the situation seen in North Africa as indicated by remains from Dar-es-Soltane and Temara, and maybe also in southern China at Zhirendong. Alternatively, East Asia may have been colonised during multiple waves during the Pleistocene, with the Longlin-Maludong morphology possibly reflecting deep population substructure in Africa prior to modern humans dispersing into Eurasia.

Citation: Curnoe D, Xueping J, Herries AIR, Kanning B, Tac¸on PSC, et al. (2012) Human Remains from the Pleistocene-Holocene Transition of Southwest China Suggest a Complex Evolutionary History for East Asians. PLoS ONE 7(3): e31918. doi:10.1371/journal.pone.0031918 Editor: David Caramelli, University of Florence, Received December 19, 2011; Accepted January 20, 2012; Published March 14, 2012 Copyright: ß 2012 Curnoe et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This research was funded by the Australian Research Council. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] (DC); [email protected] (JX)

Introduction inadequate dating [1,2–3]. For clarity, we consider East Asia to comprise the geographic region bordered by the Ural Mountains Research about the evolution of modern humans has histori- in the west, the Himalayan Plateau in the southwest, Bering Strait cally focused on the fossil records of Europe and Africa as well as in the northeast, and extending into island . the Levantine corridor connecting them. As a result, the role of the One widely discussed candidate for the oldest modern human in vast Asian continent in this evolutionary episode remains largely East Asia is the skeleton from Liujiang, southern China [2]. Yet, unknown. Human remains from the Upper Pleistocene of South the geological age of this individual has ‘‘been an everlasting Asia are scarce, being confined to just two sites possibly within the dispute since the discovery of the fossils in 1958’’ as ‘‘there is no 33-25 thousand years (or ka) range [1]. In East Asia, human fossils documentation on the exact stratigraphic position of the human are more numerous [2], but their significance has been difficult to remains’’ [4, p. 62]. As a result, its estimated age lies within the assess due to poor knowledge of their geological context and broad range of .153-30 ka [2,4]. The age of the Upper Cave

PLoS ONE | www.plosone.org 1 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

() remains is similarly problematic and has been a lineages of modern Northeast Asians and Europeans [15]. Several major source of uncertainty since their discovery in the 1930s, with later migrations seem to have occurred within the region, some estimates ranging from ,33-10 ka [2,4]. Furthermore, the Niah associated with the [12–14]. Finally, DNA extracted from Cave child from East possesses uncertain provenience a .50 ka hominin fossil from in Central Asia [5]. However, a recent field and lab program aiming to assess the belonging within the Neandertal lineage shares features exclusively stratigraphy and dating of the deposits at the site has proposed an with Aboriginal Southeast Asians and Australasians [16–18]. This age of ,45-39 ka for this cranium [5]. has been interpreted as: 1) evidence for interbreeding between the Most other candidates for the earliest modern humans in East ‘Denisovans’ and the earliest modern humans to colonise the region; Asia are similarly problematic. Among the human remains and 2) implying occupation of Southeast Asia by this archaic recovered from Tabon Cave, , the only taxonomically population during the Upper Pleistocene [16–17]. diagnostic specimen is a frontal bone assigned to H. sapiens [6], and Given the central importance of the East Asian fossil record to dated to 16.562 ka [7]. Moreover, the oldest specimen from the site testing regional and global scenarios of human evolution, in 2008 – directly dated to 47+11/210 ka [7] – might be from an we began a collaborative research project with the aim of orangutan [6]. At , , a hominin metatarsal has determining the age and providing detailed comparisons of possible been directly dated to an estimated 66.761 ka [8]. This specimen is, Pleistocene human remains from southwest China. This paper however, difficult to classify reliably, making its assignment to H. focuses on human remains from two localities: Longlin Cave sapiens uncertain [8]. A recently described individual from Tianyuan (Longlin or LL) and Malu Cave (Maludong or MLDG) (Figure 1). Cave near Zhoukoudian town, northeast China, is estimated to be The Longlin human remains were discovered opportunistically in ,42-39 ka [9]. The Tianyuan partial skeleton comprises 34 pieces 1979 by a petroleum geologist (Li Changqing) in a cave near De’e, apparently from the same individual, its femur being directly dated Longlin County, Guangxi Zhuang Autonomous Region, Guangxi to 40,3286816 cal. yr BP [9]. This specimen seems to provide the Province (Figure 1). A block of consolidated fine-grained sediment best candidate for the earliest modern human in East Asia, but is containing the human remains, unidentifiable animal bones, charcoal significantly younger (.20 kya) than genetic clock estimates for and burnt clay fragments was removed from the cave and taken to colonisation of the region (see below). Finally, a mandibular Kunming in neighbouring Yunnan Province shortly after its fragment from Zhirendong, southern China, has been dated on discovery. A partial mandible and some fragments of postcranial stratigraphic grounds to .100 ka [10]. Unfortunately, the specimen bone were prepared from the block at this time [2], although, the is fragmentary and possesses a mosaic of archaic and modern remainder of the skull and other postcranial bones were only characters also making its taxonomic status unclear [10–11]. prepared from the sediment by our team during 2010. During Given ongoing uncertainty surrounding the human fossil record, preparation we recovered a thin adhering to the surface of palaeoanthropologists have come to rely on the results of genetic the vault of the partial LL 1 skeleton, while charcoal fragments were sequencing of samples from living populations to reconstruct the collected from sediment within its endocranial cavity. Association of origins of modern humans in East Asia. Genetic research suggests cranium, mandible and postcranial elements with similar preserva- that the earliest humans dispersed into Eurasia from Africa around tion from within a small (,1m3) block of sediment suggests that post- 70-60 ka, rapidly colonising Southeast Asia and Australasia after depositional disturbance was limited. The cave has been closed to the this time [12–15]. This seems to have been followed by a later public and we have so far been unable to undertake research to clarify migration within Eurasia after 40-30 ka, adding the founding the stratigraphy and geological context of the human remains.

Figure 1. Geographic location of Longlin and Maludong in relation to major later Pleistocene human fossil sites in mainland China. 1 = Lijiang; 2 = Longtanshan; 3 = Zhirendong; 4 = Liujiang; 5 = Maba; 6 = Ziyang; 7 = Huanglong; 8 = Salawusu; 9 = Xujiayao; 10 = Zhuoukoudian-Upper Cave. doi:10.1371/journal.pone.0031918.g001

PLoS ONE | www.plosone.org 2 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Maludong is a partially mined cave fill located near the city of remaining ,3.7 m section at the site we identified 11 distinct Mengzi, Honghe Prefecture, Hani and Yi Autonomous Region, stratigraphic aggregates (Figure 2). AMS radiocarbon ages from 14 southeast Yunnan Province [2,19] (Figure 1). The site was originally charcoal samples were used to determine an age versus depth excavated in 1989 by a Chinese team including one of us (BZ), and profile (Table 1, Figure 2; see also, Table S1), providing most of the fossil and archaeological materials were recovered at that unambiguous absolute ages for the stratigraphic units recognised time [19]. Excavations during 2008 by several of the present authors at Maludong. Radiocarbon dating of bone was unsuccessful due to (DC, JX, AH, BK, BZ, ZY & LY) allowed for a re-evaluation of the a lack of preserved collagen. A magnetic susceptibility record remaining stratigraphic section and the collection of a large number corroborates the stratigraphically coherent and internally consis- of samples for dating and archaeomagnetic analysis. Additional tent radiocarbon-based chronology for the site, indicating that the human remains were discovered during the current study, both dated charcoal was deposited at the same time as its enveloping during our small-scale excavation (506506370 cm) for stratigraphic sediments (Figure 2; see also, Text S1). analysis and from unstudied and unsorted fossil collections recovered Calibrated radiocarbon ages show that the entire sequence during the 1989 field season. spans the interval 17,8306240 cal. yr BP (OZM152) to 13,2906125 cal. yr BP (OZM870) (Table 1). All of the human Results remains were recovered from within a series of deposits dating from 14,3106340 cal. yr BP (OZM149; 292 cm depth) to Dating analyses 13,5906160 cal. yr BP (OZM145; 166 cm depth), a period of Radiocarbon dating of charcoal from sediment removed from about 720 years. Moreover, the high fine-grained ferrimagnetic within the endocranial cavity of LL 1 provided an age of content of the deposits (Text S1), with their high magnetic 11,5106255 cal. yr BP (OZM369) (Table 1; see also, Table S1). susceptibility, suggests these were formed under warm, wet Three U-Th age determinations were attempted on ,25 mg sub- conditions, consistent with the Bølling-Allerød interstadial samples of the flowstone attached to the LL 1 vault (Table 2). Two (,14.7-12.6 ka [20]). Human remains recovered in situ during of these were too contaminated with detrital Th from cave the 2008 excavation and a reasonably complete calotte (specimen sediments to allow calculation of useable age estimates, but were MLDG 1704) derived from a subsection of these deposits dated able to be used to derive a robust estimate of initial 230Th/232Th between 13,9906165 cal. yr BP (OZM148; 235 cm) and activity in this contaminating phase (0.8260.20). The remaining 13,8906140 cal. yr BP (OZM146; 200 cm) (Figure 2). less-contaminated age determination has been corrected using this figure to provide an absolute age of 7.860.5 ka (UMB03650) for the Morphological description and comparison flowstone (Table 2). The flowstone must have formed after the A full list of human remains recovered from Longlin and skeleton was deposited, but its dating confirms the Pleistocene- Maludong is provided in Table 3. Here we describe and compare Holocene transition age based on radiocarbon dating of charcoal. the cranial, mandibular and dental remains, as they are the most During the original excavation at Maludong three stratigraphic informative with respect to evolution and systematics. Details of layers were identified [19]. However, in our recent research on the comparative samples are provided in Table 4 (see also data

Table 1. Radiocarbon age calculations (arranged by depth: see also, Table S1).{

Lab code Sample code Stratigraphic Conventional Radiocarbon Mean Calibrated Depth

Unit Radiocarbon Age Age Error Age (Years BP) (±1s, years) (years BP, ±2s error) (m)

Longlin ANSTO OZM369 - 10014 64 115106255 - Maludong ANSTO OZM870 BRRS 11425 50 132906125 20.192 ANSTO OZM143 RS 11527 51 133806125 20.737 ANSTO OZM154 RS 11675 52 135406165 20.894 ANSTO OZM144 DGCA 11874 49 137206150 21.200 ANSTO OZM145 ORS 11749 49 135906160 21.660 ANSTO OZM146 GAS 12037 54 138906140 21.995 ANSTO OZM147 GAS 11982 78 138406200 22.001 ANSTO OZM155 GAS 12020 54 138806140 22.020 ANSTO OZM148 GAS 12137 51 139906165 22.348 ANSTO OZM153 GAS 12430 57 145606425 22.398 ANSTO OZM149 ALROC 12304 59 143106340 22.919 ANSTO OZM150 OROC 13490 65 166306270 23.313 ANSTO OZM151 LGAC 13683 62 168206185 23.500 ANSTO OZM152 BASE 14699 65 178306240 23.900

{BP = Before Present (defined as 1950). doi:10.1371/journal.pone.0031918.t001

PLoS ONE | www.plosone.org 3 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 2. Uranium series results and age calculations."

Lab No. & 21 230 238 234 238 * 232 238 230 232 Age 234 238 { Sample U (ngg )[Th/ U] [ U/ U] [ Th/ U] [ Th/ Th] [ U/ U]i Date (ka){

Longlin-1 UMB03649 165 0.775(09) 1.471(05) 1.0317(281) 0.8 215(42) 1.453(51) Oct-2010 Longlin-2 UMB03650 94 0.146(04) 1.758(08) 0.0321(004) 4.5 7.8(0.5) 1.775(08) Oct-2010 Longlin-3 UMB03651 236 1.446(11) 1.241(05) 1.5956(180) 0.9 - - Oct-2010

"Numbers in brackets are 95% uncertainties of the given least significant figures. *Activity ratios determined after Hellstrom [53] using the decay constants of Cheng et al. [54]. 230 230 232 {Age in kyr before present corrected for intial Th using eqn. 1 of Hellstrom [55] with [ Th/ Th]i of 0.8260.20 for Longlin. {Initial [234U/238U] calculated using corrected age. doi:10.1371/journal.pone.0031918.t002 sources, Text S2). The grade term ‘modern’ is used interchange- sphenoids, anterior occipital, including anterior margin of the ably with H. sapiens sensu stricto (i.e. beginning with Omo-Kibish 1, foramen magnum, partial left occipital condyle and basioccipital Herto and other so-called anatomically modern humans to recent clivus remain. The temporal fragment (Figure 4) preserves a humans), while ‘archaic’ grade refers to all other hominin section of the squama, the base of the mastoid process (tip broken specimens/taxa. away), tympanic part with a damaged external acoustic meatus, Preservation. The LL 1 partial skull (Figure 3) preserves a mostly complete and pathologically unmodified mandibular fossa, mostly complete frontal squama with left supraorbital margin, but base of the styloid process, vaginal process and stylomastoid lacks the right lateral supraorbital part and zygomatic process. The foramen, large carotid canal, preserved foramen lacerum, foramen superior section of the nasal bones and superomedial orbital walls ovale and foramen spinosum, and a largely intact petrous part. are present, as are the left and right frontal processes of the The MLDG 1704 calotte (Figure 5) comprises mostly complete maxillae. Most of the left facial skeleton is present and comprises a frontal and paired parietal bones, but lacks its occipital, temporals nearly complete zygomatic process, alveolar process from mid-line and most of the sphenoids, as well as the entire viscerocranium. to M1, and a largely intact left zygomatic. The right maxilla is Evidently the specimen lost its base and facial skeleton owing to incomplete save much of the lateral margin of the piriform anthropogenic alteration, with cut-marks seen along the walls of aperture and alveolar process. What remains has been rotated the vault and on the zygomatic process. Its preserved morphology ,45u from the median sagittal plane owing to post-burial is unaffected by this alteration. The specimen is free from post- compression. The left side (MSP to lateral) is largely free of deposition distortion as indicated by visual inspection and scrutiny distortion, with the landmarks prosthion and nasospinale readily of CT-scans. identifiable. The tip of the anterior nasal spine is broken away, but The LL 1 mandible (Figure 6) and two partial mandibles its base is easily discerned. The bony palate lacks most of the left recovered from Maludong (MLDG 1679 and MLDG 1706: and right palatine processes. The morphology of the preserved left Figure 7) are also compared. Specimen LL 1 comprises a largely maxillary tuberosity is consistent with M3 agenesis. Parts of the complete body, but is missing its left ramus save the root and

Figure 2. Maludong site. (A) stratigraphic sequence; (B) GIS plotted stratigraphy based on total station data and indicating excavation units and plotted finds; (C) stratigraphic aggregates; (D) Upper and Lower age limit of calibrated radiocarbon ages; and (E) magnetic susceptibility record (61026 m3 kg21). doi:10.1371/journal.pone.0031918.g002

PLoS ONE | www.plosone.org 4 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 3. Human remains recovered from Longlin and Maludong.

Locality Catalogue No. Description

Longlin LL 1 1 2 1 1 Partial cranium with left I -M ; right I , fragment of M , partial mandible with left I1, canine, P3,M2-M3; right I1-I2, canine, P3-P4 and M2, several isolated tooth fragments, almost complete left half of axis (vertebra C2), proximal ulna fragment and several rib fragments Maludong MLDG 1678 Femur, proximal ,Mrd

1679 Mandible, right body fragment and ramus, M2 and M3 crowns 1704 Cranium, calotte 1705 Cranial vault fragments 1706 Mandible, right hemi-mandible, no dentition 1707 Manual phalanx 1708 Cranium, zygomatic fragment 1710 Ulna, proximal fragment 1711 Manual phalanx, proximal fragment 1712 Rib fragment 1713 Cranium, maxillary body fragment, no dentition 1714 Rib fragment 1715 Rib fragment 1716 Rib fragment 1717 Femur, ,K head 1718 Sacrum, partial 1722–1730 Parietal fragments 1731–1733 Frontal fragments 1734 Occipital fragment 1735–1739 Parietal fragments 1740 Occipital fragment 1741 Parietal fragment 1742–1744 Parietal fragments 1745 Occipital fragment 1746 Frontal fragment 1747 Left M3 1748 Left upper partial premolar (P3) 1749–1750 Sternum, ,K of manubrium and ,K of body 1751 Left partial I2 1756 Sacrum, partial

doi:10.1371/journal.pone.0031918.t003 coronoid process, and lacks the entire right ramus. The position of Specimen MLDG 1679 (Figure 7) comprises a right mandibular the take-off of the left ramus relative to M3 makes clear that a body fragment preserved from just anterior to M2, with intact M2- retromolar space would have been present (M3 being uncovered M3 crowns and roots, and including a complete ramus in two [21]). The external surface of the symphysis has been displaced pieces. The internal morphology of the body and ramus is well superiorly such that the bone is out of alignment with the alveolar preserved including the mandibular foramen, pterygoid surface, process. This makes accurate assessment of chin development and coronoid and condylar parts (the former having been modified problematic. The left alveolar part retains the roots and crowns of somewhat by osteoarthritis). I1, canine, P3, partial P4,M2 and partial M3. The first molar is Specimen MLDG 1706 (Figure 7) is a right hemi-mandible, missing and the alveolar bone shows signs of ante-mortem tooth broken just lateral to the symphysis (left side), through to a mostly loss with resorption and new bone growth/remodelling. Much of complete ramus. The body is damaged (abraded) along its inferior the right body is preserved and retains the mental foramen, I1-P4 border, while scoring marks the external surface of the symphysis. and M2 roots and crowns. The right M1 seems again to have been No dental crowns were recovered with the specimen, but all of the lost ante-mortem, with signs of remodelling of the alveolar bone. alveoli are open and lack signs of bony remodelling, indicating a The transverse tori are somewhat thickened such that the internal full (adult) set of dentition would probably have been present about surface of the symphysis is not vertical, a small internal plane being the time of death. Externally, the mental foramen is present and present. well preserved. Internally, the morphology of the surface of the

PLoS ONE | www.plosone.org 5 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 4. Cranial series employed in comparisons (where data were compiled by the present authors: see Text S2 for a list of data sources for metrical and morphological data and dating estimates).

Sample Taxon & Sample Site/series Estimated Region Abbrev. Age (kya)

Early Homo sapiens East Asian EAEHS Baojiyan, Chuandong, Dhongzhongyan, Du’an, Gua Gunung, Hang Cho, ,67-10 Huanglong PA 842, Liujiang, Mai Da Nuoc, Minatogawa 1, 4, Moh Khiew, Tubo, Nalai, Upper Cave Zhoukoudian 101, 103, Wushan, Wadjak European EUEHS Barma Grande unnumbered, 1, 2, Combe Capelle, Cotte de St. Brelade, ,32-.20 Cro Magnon 1, 2, 3, Mladec 1, 2, 5, 6, Predmost 1, 3, 4, 9, 10, 14, Grotte de Enfants 4, 5, 6 West Asian WAEHS Skhul 2, 4, 5, 6, 9, Qafzeh 1, 2, 3, 5, 6, 7, 9 ,173-36 African AFEHS Herto BOU-VP-16/1 and Omo-Kibish 1 ,195-150 Homo neanderthalensis NEAND Amud 1, Arcy-sur-Cure, Ehringsdorf 1, 2, Forbe’s Quarry, 1, Guattari 1, .250-,32 Krapina C, D, E, 3, 4, 6, 16, 27/28, 34.1, Krapina Par. 5, Par. 20, Par. 21, Par. 32, Kulna, La Chapelle, 1, , 5, 13, Le Moustier, Monte Circeo, Neandertal, Saccopastore 1, Sal’a 1, Shanidar 1, 2, 4, 5, Spy 1–3, Tabun 1, Vindija VI 204, 261, 293, 284–230–255–256, VI 224, VI 227, VI 261 East Asian Middle Pleistocene archaic Hominins EAMPH Dali, Jinniushan, Maba, Xujiayao ,195-.104 Homo erectus sensu stricto ERECT Buku, Chenjiawo, Gongwangling, Hexian, Jianshi PA504, PA502 and PA503, ,1600-143 Luonan, Nanjing, Ngandong 1, 3, 5, 6, 7, 9, 10, 11, 12, 14, Sangiran 1b, 2, 3, 4, 9, 10, 12, 17, 18, 21, 22, 38, Sb 8103, T2, Sumbangmacan 1, 3, 4, Trinil II, Wushan, Yiyuan, Yunxian, Xichuan, Zhoukoudian 1, 2, 3, 4, 5, 6, 10–13, 18, 20–33, 43–52, 69, 70–74, 80–95, 106–117, 131, 134, 136, 138, 140, 143–144, 146, Zhoukoudian reconstruction Howells samples East Asian Ainu, Andaman, Atayal, Buriat, , Hainan, N , Philippines, Recent S Japan Australian Australia, Tasmania Recent European Berg, Norse, Zalava´r Recent African Bushman, Dogon, Teita, Zulu Recent

doi:10.1371/journal.pone.0031918.t004 body and ramus is clear: the symphysis expresses enlarged tori, the (2062 mm), the difference from LL 1 being significant (z-score mandibular foramen is clear, and the pterygoid surface and 24.27, p0.001; MLDG 1704 z-1.42). At mid-orbit [24], coronoid and condylar parts are well preserved (the former being supraorbital projection is comparatively weak in LL 1 (11 mm; modified slightly by osteoarthritis). EUEHS z-1.58), but moderate in MLDG 1704 (16 mm), and Supraorbital region. The supraorbital part of LL 1 is identical to the EUEHS mean (Table 5). In contrast, mid-orbit conspicuous, with a well-developed glabella. It lacks the obvious projection [24] is strong in WAEHS (20 mm) and NEAND signs of division typically seen among modern humans (i.e. lacks a (2262 mm; MLDG 1704 z-2.97, p0.002; LL 1 z-5.44, p,0.0001). dividing sulcus between medial and lateral parts), but it does thin At the lateral position [24], projection in LL 1 is similar to EUEHS in the vertical dimension mediolaterally. The supraorbital torus of (2063 mm; z-0.32), while in MLDG 1704 it is strong (,23 mm; MLDG 1704 is marked by a strongly developed glabella and EUEHS z0.95), the specimen closely resembling WAEHS superciliary ridges, but also thins laterally. Its supraorbital part is, (24 mm, n4) and NEAND (2462 mm; z-0.49; LL 1 z-2.47, however, divided into medial and lateral components by a distinct p0.009). sulcus, being bipartite in form. The presence of a supraorbital Vertically thickness of the supraorbital at the medial location torus is a condition rarely seen in recent humans [22], but is more [24] is similar in LL 1 (15 mm) and MLDG 1704 (L 16.5/R frequent among Pleistocene H. sapiens [23]. A bipartite 17 mm) to EUEHS (1763 mm; LL 1 z0.64; MLDG 1704 z-0.08) supraorbital like that seen in MLDG 1704 is characteristic of H. and NEAND (1763 mm; LL 1 z-0.63; MLDG 1704 z-0.08). sapiens and distinguishes it from archaic taxa [23]. Their values are, however, slightly reduced compared to WAEHS Table 5 compares supraorbital projection and vertical thickness (1863 mm; z-0.94, z-0.39). Mid-orbit thickness [24] is moderate dimensions. Supraorbital projection at the medial location [24] is in LL 1 (7 mm) and similar to WAEHS (863 mm; z-0.31), but moderate in LL 1 (11 mm), but high in MLDG 1704 (17 mm, well below the NEAND mean (1062 mm; z-1.48). In MLDG measured in MLDG 1704 on CT-scans). Longlin resembles 1704, thickness at this location is marked (10/13 mm), and while European early H. sapiens (or EUEHS) in this regard (1363 mm; exceeding mean values for all comparative samples, it is most z-score adjusted for the size of the comparison sample [25] 20.63), similar to NEAND (z0.74; contrasting with EUEHS z3.11, p0.005; while MLDG 1704 is identical to West Asian early H. sapiens (or and WAEHS z1.09). Finally, vertical thickness at the lateral WAEHS) crania (i.e. Skhul and Qafzeh: mean 17 mm). Values for location [24] is comparatively thin in LL 1 (7 mm) and MLDG both specimens are below the H. neanderthalensis (or NEAND) mean 1704 (7/6.5 mm). They both closely resemble the laterally thin

PLoS ONE | www.plosone.org 6 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Figure 3. Longlin 1 partial skull (each bar = 1 cm). doi:10.1371/journal.pone.0031918.g003 supraorbitals of EUEHS (861 mm; LL 1 z-0.96; MLDG 1704 z- (archaic) hominins (or EAMPH) and ERECT (both 1062 mm; z- 1.20) and WAEHS (1063; LL 1 z-0.94; MLDG 1704 z-1.01). In 1.37 and z-1.47). contrast, NEAND tori are laterally thick (1262 mm; LL 1 z-2.47, Vault dimensions. Tables 6–7 compare a range of vault p0.008; MLDG 1704 z-2.59, p0.006). measurements for LL 1 and MLDG 1704 with various Pleistocene Vault thickness. Vault thickness measurements are modern (Table 6) and archaic hominin (Table 7) samples. A presented for LL 1 and MLDG 1704 and compared in Table 5. comparison of African early H. sapiens (or AFEHS) and NEAND At bregma, LL 1 has a thick vault (10 mm), being most similar to helps to sort the polarities of features found in Eurasian samples. the H. erectus (ERECT) mean (962 mm; z0.49). While its value is The Herto (BOU-VP-16/1) cranium possesses a large endocranial within one standard deviation unit of the EUEHS sample mean volume (ECV) (1450 cm3). It is, however, only slightly enlarged (763 mm; z0.94), it is significantly different to the NEAND mean compared with NEAND (14076172 cm3). In contrast, the greatly (761 mm; z2.89, p0.006). Thickness at bregma in MLDG 1704 enlarged ECV of WAEHS (Skhul-Qafzeh: 1556625 cm3)is (7 mm) is identical to mean values for EUEHS, WAEHS and significantly larger than Herto (z-3.87, p0.008) and NEAND (t- NEAND (Table 5). At the parietal eminence, thickness in MLDG 5.44, p0.002), indicating that ECV enlargement is a derived 1704 (7.6/6.4 mm) is within one standard deviation unit of means characteristic of Pleistocene Eurasian H. sapiens. Reconstructed for EUEHS (661 mm, z0.97) and WAEHS (both 862 mm, z- ECV for MLDG 1704 (,1327 cm3: Figures S1, S2, S3, S4) is 0.47), but distinct from the means of East Asian Middle Pleistocene small in comparison with all H. sapiens sample means: its value sits

PLoS ONE | www.plosone.org 7 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Figure 5. Maludong 1704 calotte (each bar = 1 cm). doi:10.1371/journal.pone.0031918.g005 Figure 4. Longlin 1 temporal fragment (each bar = 1 cm). doi:10.1371/journal.pone.0031918.g004 is moderate in AFEHS and EAEHS (120 mm), and WAEHS (11964 mm) and NEAND (12266 mm). MLDG 1704 possesses a broad maximum frontal width (125 mm), in common with outside of (below) the ranges of WAEHS (z-8.36, p0.0005) and EUEHS (12467 mm, z0.14). EUEHS (z-1.71, p0.05). ECV for East Asian early H. sapiens (or The parietal bones of AFEHS are long (chord 125 mm, arc EAEHS) (14076146 cm3; MLDG z-0.51) is identical to the 135 mm) and contrast strongly with the short parietals of NEAND NEAND mean (Table 7). Among all samples, MLDG 1704 is (chord 10864 mm, arc 11565 mm). The parietals of MLDG closest to these latter sample means. Thus, MLDG 1704 shares 1704 are short (chord 107 mm, arc 123 mm) by Pleistocene H. with EAEHS and NEAND a reduced ECV. This reduction sapiens standards (Table 6). Its parietal chord is significantly shorter contrasts late Pleistocene humans in East Asia with earlier than the EAEHS (11764 mm; z-2.34, p0.02) and EUEHS Eurasian and African modern humans with their large ECVs. (12067 mm; z-1.81, p0.04), and most closely resembles NEAND The frontal bone of the earliest modern humans is long (AFEHS (z-0.24). Its arc dimension is, however, closer to H. sapiens (means frontal chord 124-131 mm, arc 153 mm) and contrasts with the 132–135 mm; MLDG 1704 z-0.94 to z-1.19) than to archaic short frontals of NEAND (chord 11266 mm, arc 12369 mm). species (means 106–117.5 mm; MLDG 1704 z1.17 to z3.33; with The frontal bones of LL 1 (chord 112 mm, arc 134 mm) and p0.001 for ERECT). Broad parietals are characteristic of NEAND MLDG 1704 (chord 116 mm, arc 133 mm) are moderate in (14867 mm), distinguishing them from H. sapiens (means 138– length, being similar to EUEHS and ERECT, but distinguishable 145 mm) and ERECT (14266 mm). The absolutely narrow bi- from the very long frontals of AFEHS. Minimum frontal breadth parietal breadth of MLDG 1704 (141 mm) is most similar to the is wide in AFEHS (i.e. Herto 112 mm), contrasting with the mean of EUEHS (14367 mm, z-0.28). narrower post-orbitals of NEAND (10365 mm). This region is The ratio of parietal/frontal chord and parietal/frontal arc even narrower in LL 1 (94 mm) and MLDG 1704 (95 mm), distinguishes samples of Eurasian H. sapiens (means: chord 104– contrasting with the moderately broad anterior frontals of EAEHS 107%, arc 99–107%) from NEAND (chord 9868%, arc (9965 mm), EUEHS (10165 mm; z-1.36, z-1.17), and WAEHS 96610%). The shortened parietals of AFEHS (chord 98%, arc and NEAND (both 10365 mm; WAEHS z-1.67, z-1.48; NEAND 85%) are, in contrast, a putative ancestral trait shared with z-1.76, p0.04, z-1.56, p0.06). The LL 1 and MLDG 1704 values NEAND. Thus, parietal sagittal expansion is characteristic of are most similar to the mean for ERECT (93610 mm; z0.10, Pleistocene Eurasian H. sapiens. For these indices, MLDG 1704 z0.20). Maximum frontal breadth, measured at the coronal suture, (both 92%) is highly distinct from H. sapiens (EAEHS z2.24, p0.03;

PLoS ONE | www.plosone.org 8 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Figure 7. Maludong mandibles MLDG 1679 (left) and MLDG 1706 (right) (scale bar = 1 cm). NB: MLDG 1706 is broken through its symphysis just lateral to the MSP. doi:10.1371/journal.pone.0031918.g007

Mandibular fossa. Measurements of the left mandibular fossa of LL 1 (Figure 4) are compared in Table 8. Its mandibular fossa is moderate in length (A-P 17 mm), being similar to the WAEHS (Skhul-Qafzeh: 19 mm), ERECT (2062 mm; z-1.04) and EUEHS (21 mm) means. It is, however, significantly different to the means of recent Africans (1161 mm; z5.97, p,0.00001), a sample of Pleistocene early H. sapiens (comprising Ngaloba, 1 and 2, Singa 1, Omo-Kibish 2, Skhul 4 and 5 [26]) (1162 mm; z2.81, p0.01) and NEAND (1161 mm; z5.74, p,0.00001). The mandibular fossa of LL 1 is very broad (M-L c31 mm) and lies outside of the range of all comparative samples, being closest to EAMPH (30 mm). Its breadth is significantly different to the means of recent Africans (2062 mm; z5.47, p,0.00001), Pleistocene early H. sapiens (composition, see above: 2362 mm; z3.74, p0.004) and NEAND (2163 mm; z3.19, p0.004). Its mandibular fossa is comparatively deep (S-I 13 mm) like ERECT (1264 mm; z0.23), but shallower than EAMPH (17 mm). In contrast, samples of H. sapiens exhibit smaller mean values or much shallower fossae (5–6 mm; z6.55–7.96; p0.0003- ,0.00001). Shallow fossae are also characteristic of NEAND (662 mm; z3.35, p0.003). Facial skeleton. The facial skeleton of LL 1 is unusual compared with early H. sapiens in exhibiting strong alveolar prognathism. The mid-face is flat, both at the nasal root and piriform aperture, and zygomatic process of the maxilla. The specimen lacks a canine fossa, but possesses a deep sulcus maxillaris. The left zygomatic arch is laterally flared. The zygomatic bone is strongly angled such that its inferior margin sits well lateral to the Figure 6. Longlin 1 mandible (scale bar = 1 cm). superior part. The zygomatic tubercle is small and sits lateral to a doi:10.1371/journal.pone.0031918.g006 vertical line projected from the orbital pillar. The anterior section of the masseter attachment is marked by a broad and deep sulcus, EUHS z1.67, p0.05), and most closely resembles archaic hominins but the zygomatic tubercle is small. The anterior wall of the such as East Asian Middle Pleistocene hominins (or EAMPH: zygomaticoalveolar root is in an anterior position (above P4/M1). chord index 92%), ERECT (chord: 8767%, z0.41) and NEAND The lateral orbital margin (pillar) exhibits strong transverse (arc: 96610%, z-0.38). incurvation when viewed in lateral aspect. In most of these A commonly deployed index of postorbital width is the frontal features, LL 1 displays the putative plesiomorphic condition for constriction index, or ratio of minimum/maximum frontal later hominins, being highly distinguishable from the modal breadth. Its value for MLDG 1704 (76%) is unusually low, and condition of H. sapiens. while it sits (just) within the lower part of the range of EAEHS (76– Tables 9–10 compare standard measurements and indices of 89%), it is most similar to the mean for ERECT (8265%, MLDG the facial skeleton for LL 1 and a single measurement for MLDG z-1.17). In contrast, MLDG 1704 is distinct from mean values for 1704 with Pleistocene modern human (Table 9) and archaic EUEHS (8262%; z-2.91, p0.005), WAEHS (8363%; z-3.04, (Table 10) samples. Data for superior facial breadth are p0.01) and NEAND (8665%; z1.93, p0.03). unavailable for AFEHS. However, the narrower upper face of

PLoS ONE | www.plosone.org 9 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 5. Bone thickness measurements compared (significant z-scores in bold)."

LL MLDG EUEHS WAEHS NEAND EAMPH ERECT 1 1704

Supraorbital projection 1. Medial 11.0 17.0 1363(9) 17(4) 2062(9) - - (8–18) (11–21) (18–23) z-score/p LL 1 20.63/0.27 - 24.27/0.001 -- z-score/p MLDG 1704 1.26/0.13 - 21.42/0.09 - - 2. Midorbit 11.0 16.0 1663(9) 20(4) 2262(43) - - (8–19) (16–25) (17–25) z-score/p LL 1 21.58/0.07 - 25.44/,0.00001 -- z-score/p MLDG 1704 n.a. - 22.97/0.002 -- 3. Lateral 19.0 23.0 2063(9) 24(4) 2462(35) - - (15–23) (21–26) (21–25) z-score/p LL 1 20.32/0.37 - 22.47/0.009 -- z-score/p MLDG 1704 0.95/0.18 - 20.49/0.31 - - Supraorbital vertical thickness -- 4. Medial 15.0 16.5/17.0* 1763(11) 1863(7) 1763(9) - - (12–24) (14–22) (15–22) z-score/p LL 1 20.64/0.26 20.94/0.19 20.63/0.27 - - z-score/p MLDG 1704 20.08/0.46 20.39/0.35 20.08/0.46 - - 5. Midorbit 7.0 10.0/13.0* 562(11) 863(7) 1062(48) - - (6–10) (5–11) (8–12) z-score/p LL 1 0.96/0.18 20.31/0.38 21.48/0.07 - - z-score/p MLDG 1704 3.11/0.005 1.09/0.15 0.74/0.23 - - 6. Lateral 7.0 7.0/6.5* 861(11) 1063(7) 1262(42) - - (6–10) (6–14) (10–14) z-score/p LL 1 20.96/0.18 20.94/0.19 22.47/0.008 -- z-score/p MLDG 1704 21.20/0.12 21.01/0.17 22.59/0.006 -- Vault thickness 7. Bregma 10.0 7.0 763(8) 7(2) 761(13) 8(4) 962(22) (4–12) (6–8) (4–9) (7–9) (6–16) z-score/p LL 1 0.94/0.18 - 2.89/0.006 - 0.49/0.31 z-score/p MLDG 1704 n.a. - n.a. - 20.98/0.16 8. Parietal eminence - 7.6/6.4* 661(16) 862(9) 862(30) 1062(5) 1062(16) (4–10) (5–11) (5–17) (7–13) (5–16) z-score/p LL1 --- -- z-score/p MLDG 1704 0.97/0.17 20.47/0.32 20.47/0.32 21.37/0.12 21.46/0.08

"Above the line m6s(n), below the line (min.-max.); z-tests corrected for small comparative sample size; Bonferroni correction not employed as per [56]; sample abbreviations and compositions see Table 4; data sources see Text S2. *Mean of left and right used in z-test. doi:10.1371/journal.pone.0031918.t005

NEAND (11864 mm) distinguishes them from the very broad The facial skeleton of LL 1 is broad. Bizygomatic breadth is upper facial skeletons of WAEHS (Qafzeh-Skhul: 12367 mm). estimated to be wide (c144 mm), strongly distinguishing it from This later morphology is shared by WAEHS with ERECT EAEHS, the value for LL 1 being outside of (slightly above) its (123 mm). Contrasting with both of these conditions are later range. Its bizygomatic most closely resembles NEAND Eurasian samples of H. sapiens with their markedly narrow superior (14568 mm; z-0.12), and is similar also to AFEHS (142 mm). A facial skeletons (EAEHS and EUEHS: mean 112 mm). A narrow second index of postorbital constriction is the ratio minimum superior facial breadth is a shared condition of LL 1 (c110 mm) frontal breadth/bizygomatic breadth, providing a more direct and MLDG 1704 (109 mm), both with each other and with later measure of the relative size of the temporal fossa. The value for LL Pleistocene Eurasians. Their values are, however, highly distinct 1 is large (66%) by later hominin standards. While it is equal to the from WAEHS (LL 1 z-1.72, p0.07; MLDG 1704 z-1.85, p0.06) minimum value for EUEHS and WAEHS, its value is distant from and NEAND (LL 1 z-1.96, p0.03; MLDG 1704 z-2.21, p0.01). their means (EUEHS (7364%; z-1.67, p0.06; WAEHS 70%). It

PLoS ONE | www.plosone.org 10 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 6. Vault measurements (cm3, mm) and indices (%) compared to Pleistocene modern humans (significant z-scores in bold)."

Measurement LL MLDG EAEHS EUEHS WAEHS AFEHS1 Abbrev. (Martin No.) 1 1704

1. Endocranial volume - (1327) 14076146(7) 15666134(12) 1556625(5) 1450(1) ECV (1170–1567) (1375–1880) (1518–1587) - z-score/p MLDG 1704 20.51/0.31 21.71/0.05 28.36/0.0005 - 2. Frontal chord 112 116 11266(7) 11666(17) 113(4) 127.5(2) FRC* (M29) (105–119) (91–111) (106–118) (124–131) z-score/p LL 1 n.a. 20.65/0.26 - - z-score/p MLDG 1704 0.62/0.27 n.a. - - 3. Frontal arc 134 133 13065(8) 13567(9) 128(4) 153(1) FAA (M26) (121–136) (121–148) (118–133) - z-score/p LL 1 0.75/0.23 20.14/0.44 - - z-score/p MLDG 1704 0.57/0.29 20.27/0.39 - - 4. Minimum frontal breadth 94 95 9965(7) 10165(17) 10365(6) 112(1) MFB (M9) (95–109) (91–111) (96–110) - z-score/p MLDG 1704 20.75/0.24 21.17/0.13 21.48/0.09 - 5. Maximum frontal breadth - 125 120(4) 12467(17) 11964(5) 120(1) XFB* (M10) (112–129) (107–139) (114–125) - z-score/p MLDG 1704 - 0.14/0.44 1.37/0.12 - 6. Parietal chord - 107 11764(7) 12067(19) 121(4) 125(2) PAC* (M30) (111–122) (107–143) (112–129) (121–129) z-score/p MLDG 1704 22.34/0.02 21.81/0.04 -- 7. Parietal arc - 123 13269(8) 13469(19) 133(4) 135(2) PAA (M27) (120–150) (117–157) (120–145) (130–140) z-score/p MLDG 1704 20.94/0.18 21.19/0.12 - - 8. Maximum parietal breadth - 141 13866(7) 14367(19) 14563(7) 145(1) MPB (M8) (131–145) (131–154) (140–148) - z-score/p MLDG 1704 0.47/0.32 20.28/0.39 21.25/0.12 - 9. Parietal/frontal chord index - 92 10465(7) 10467(17) 107(4) 98(2) (M30/M29) (95–112) (94–117) (98–113) (92–104) z-score/p MLDG 1704 22.24/0.03 21.67/0.05 - - 10. Parietal/frontal arc index - 92 10268(8) 9967(19) 107(3) 85(1) (M27/M26) (94–115) (87–110) (102–111) - z-score/p MLDG 1704 21.18/0.13 20.97/0.17 - - 11. Frontal constriction index - 76 83(4) 8262(16) 8663(5) 93(1) (M9/M10) (76–89) (79–86) (80–88) - z-score/p MLDG 1704 - 22.91/0.005 23.04/0.01 -

"Fossil values in round brackets are estimates; above the line m6s(n), below the line (min.-max.); z-tests corrected for small comparative sample size; Bonferroni correction not employed as per [56]; sample abbreviations and compositions see Table 4; data sources see Text S2. 1Mostly comprises measurements of Herto BOU-VP-16/1. *Equivalent to measurements of Howells [57] and employing his abbreviation. doi:10.1371/journal.pone.0031918.t006 also contrasts strongly with EAEHS (7363%; z-2.16, p0.04) and shortening is also seen in archaic EAMPH (74 mm), but not to NEAND (7462%; z-3.70, p0.006). In contrast, bimaxillary the great extent characterising late Pleistocene H. sapiens (but more breadth in LL 1 (108 mm) is most similar to EAEHS so than the AFEHS specimen from Herto). The facial index (10566 mm; z0.47). While the index of upper/mid-facial (superior height/bizygomatic breadth) for LL 1 (44%) shows its (bimaxillary) breadth is high for LL 1 (98%), it is similar to bony face to be very short relative to breadth. The value in the EAESH (9365%; z0.65) and NEAND (9565%; z0.57). specimen is not especially close to any sample mean and is Superior facial height is greatly reduced in LL 1 (64 mm), a significantly different to EAEHS (5062%; z2.81, p0.01) and condition distinguishing Eurasian early H. sapiens from AFEHS NEAND (5962%; z-6.94, p0.0004). (79 mm) and archaic hominins (means: 80.5–87 mm). The While the left orbit of LL 1 is broad (45 mm), being identical to superior facial height of LL 1 is in fact significantly shorter than the WAEHS mean, this measurement has little discriminating the mean for WAEHS (7363 mm; z-3.35, p0.01). Facial power, as LL 1’s value lies within one standard deviation unit of

PLoS ONE | www.plosone.org 11 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 7. Vault measurements (cm3, mm) and indices (%) compared to archaic hominins (significant z-scores in bold)."

Measurement LL MLDG NEAND EAMPH ERECT Abbrev. (Martin No.) 1 1704

1. Endocranial volume - (1327) 14076172{(28) 1255(2) 10286127$,¥(16) ECV (1172–1740) (1120–1390) (815–1251) z-score/p MLDG 1704 20.46/0.32 - 2.28/0.01 2. Frontal chord 112 116 11266{(18) 116(1) 11266£(12) FRC* (M29) (98–123) - (99–120) z-score/p LL 1 n.a. - n.a. z-score/p MLDG 1704 0.65/0.26 - 0.64/0.26 3. Frontal arc 134 133 12369{(12) 134(1) 12567$(12) FAA (M26) (107–135) - (110–135) z-score/p LL 1 1.17/0.13 - 1.24/0.12 z-score/p MLDG 1704 1.07/0.15 - 1.10/0.14 4. Minimum frontal breadth 94 95 10365{(21) 114(1) 93610$,¥(23) MFB (M9) (97–112) - (73–109) z-score/p MLDG 1704 21.56/0.06 - 0.20/0.42 5. Maximum frontal breadth - 125 12266{(13) - 11868$,¥(23) XFB* (M10) (108–128) - (102–123) z-score/p MLDG 1704 0.48/0.33 - 0.86/0.20 6. Parietal chord - 107 10864{(13) 110(2) 10065$,¥(21) PAC* (M30) (102–114) (107–113) (86–108) z-score/p MLDG 1704 20.24/0.40 - 1.37/0.09 7. Parietal arc - 123 11565{(10) 117.5(2) 10665$,¥(22) PAA (M27) (110–126) (114–121) (92–113) z-score/p MLDG 1704 1.53/0.08 - 3.33/0.001 8. Maximum parietal breadth - 141 14867{(16) 148.5(2) 14266¥(21) MPB (M8) (138–157) (148–149) (130–153) z-score/p MLDG 1704 20.97/0.17 - 20.16/0.43 9. Parietal/frontal chord index - 92 9868{(10) 92(1) 8967(11) (M30/M29) (89–116) - (81–104) z-score/p MLDG 1704 20.72/0.24 - 0.41/0.34 10. Parietal/frontal arc index - 92 96610(9) 85(1) 8465(11) (M27/M26) (83–115) - (71–91) z-score/p MLDG 1704 20.38/0.35 - 1.53/0.07 11. Frontal constriction index - 76 8665(13) - 8265(21) (M9/M10) (78–94) - (72–89) z-score/p MLDG 1704 21.93/0.03 - 21.17/0.12

"Fossil values in round brackets are estimates; above the line m6s(n), below the line (min.-max.); z-tests corrected for small comparative sample size; Bonferroni correction not employed as per [56]; sample abbreviations and compositions see Table 4; data sources see Text S2. *Equivalent to measurements of Howells [57] and employing his abbreviation. {T-test EUEHS and NEAND mean difference: one-tailed p,0.05-0.01. {T-test EUEHS and NEAND mean difference: one-tailed p,0.001. £T-test EUEHS and ERECT mean difference: one-tailed p,0.05-0.01. $T-test EUEHS and ERECT mean difference: one-tailed p,0.001. ¥T-test NEAND and ERECT mean difference: one-tailed p,0.001. doi:10.1371/journal.pone.0031918.t007

EAEHS (4462 mm), EUEHS (4364 mm) and NEAND orbit results in a moderate orbital index value (76%), being most (4463 mm). In contrast, orbit height is moderate in LL 1 similar to WAEMH (7467%; z0.26), but distant from NEAND (34 mm), distinguishing the specimen from the short orbits of (8766; z-1.76). EAEHS (3161 mm; z2.81, p0.01) and EUEHS (3063 mm; The piriform aperture is broad in LL 1 (maximum nasal width z1.29), and tall orbits of NEAND (3861 mm; z-3.88, p0.0007). 32 mm), being identical to the NEAND mean (Table 10). It is The value for LL 1 is identical to AFEHS and similar to WAEHS substantially broader than the means for all H. sapiens samples: (3363 mm). This combination of a broad and moderately tall EAEHS 2862mm(z1.89), EUEHS 2662mm(z2.89, p0.006)

PLoS ONE | www.plosone.org 12 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 8. Mandibular fossa dimensions (mm) of Longlin 1 (Figure 8B) shows the third principal component to distinguish compared (significant z-scores in bold)." ERECT from all other taxa. In this plot, LL 1 sits in a unique position, well away from all crania owing to a combination of a high positive score for PC 2 and moderate score for PC 3. Z-tests A-P M-L S-I of object scores indicate that the difference between LL 1 and the H. sapiens mean is not significant for all PCs (PC 1 z-0.89; PC 2 Length Breadth Depth z0.95; approaching significance for PC 3 z1.70, p0.05). LL 1 17 (31) 13 The second analysis included MLDG 1704 and 23 other crania, Recent Africans* 1161(99) 2062(99) 561(99) employed eight variables (Table 11), and generated three principal (8–14) (16–27) (4–8) components. For PC 1, the highest loading variables were parietal z-score/p 5.97/ 5.47/ 7.96/ chord and parietal arc, and these were contrasted mostly with ,0.00001 ,0.00001 ,0.00001 measures of vault width (biparietal breadth and superior facial EUEHS* 21(3) 25(3) 6(3) breadth) (Table 11). For PC 2, frontal chord and frontal arc accounted for most of the variance (Table 11), while PC 3 was (19–25) (21–30) (5–7) mostly explained by maximum frontal breadth (Table 11). A * WAEHS 19(4) 27(3) 7(4) bivariate plot of object scores for PC 1 and PC 2 (Figure 9A) shows (17–22) (26–29) (5–8) PC 1 to separate crania belonging to H. sapiens from those of Pleistocene early H. sapiens*,{ 1162(7) 2362(7) 661(7) NEAND, H. heidelbergensis sensu stricto and ERECT. Specimen (8–14) (21–25) (5–8) MLDG 1704 sits just outside of the convex hull for H. sapiens, but clusters close to Cro Magnon 1 and 3 (Figure 9A). A plot of PC 2 z-score/p 2.81/0.01 3.74/0.004 6.55/0.0003 * versus PC 3 (Figure 9B) shows that these principal components do NEAND 1161(11) 2163(11) 662(11) not discriminate well among taxa. Principal component 3 does, (9–13) (16–26) (4–8) however, distinguish MLDG 1704 from all other crania, with its z-score/p 5.74/ 3.19/0.004 3.35/0.003 high positive score. For PC 3, its score is outside of the range of all ,0.00001 crania, exceeding the next highest score by 0.29 eigenfactor units EAMPH1 27(1) 30(1) 17(1) (almost double the difference between the H. sapiens and NEAND ERECT1 2062(6) 25(4) 1264(6) means). Z-tests of object scores indicate that the difference between (17–22) (23–27) (5–15) LL 1 and the H. sapiens mean is not significant for all PCs [PC 1 z0.75; PC 2 z1.51; approaching significance for PC 3 (z1.62, z-score/p 21.04/0.10 - 0.23/0.41 p0.06)]. In contrast, the mean difference for NEAND is significant "Sample abbreviations see Table 4; LL 1 value in round brackets is an estimate; for PC 1 (z-6.89, p0.001), but not for PC 2 (z1.70) or PC 3 (z1.22). above the line m6s(n), below the line data (min.-max.); Bonferroni correction The final analysis included MLDG 1704 and 43 other crania, not employed as per [56]. employed 6 variables (Table 11), and generated two principal *Comparative sample descriptive statistics from [26]. {Composition, see text. components. For PC 1, parietal chord and parietal arc explained 1Sample compiled by the authors: compositions see Table 4, data sources see most of the variance (Table 11). For PC 2, frontal arc was Text S2. contrasted with maximum frontal breadth (Table 11). A bivariate doi:10.1371/journal.pone.0031918.t008 plot of object scores for PC 1 and PC 2 (Figure 10) shows good separation between H. sapiens on the one hand, and NEAND and and WAEHS 3061mm(z1.83). Nasal height is short (45 mm), its ERECT on the other. Specimen MLDG 1704 sits just within the value being distant from the means of early H. sapiens (means 42– H. sapiens convex hull, but near the edge of the ERECT range. It 51 mm; z1.37–1.44). It is, however, significantly different to the also sits close to the Nazlet Khater 2 cranium from , a late NEAND mean (6163 mm; z-4.94, p0.002). The nasal index for Pleistocene specimen also possessing a mix of modern and archaic LL 1 (71%) is large by later Pleistocene hominin standards, its characters [28]. The object score of MLDG 1704 for PC 1 is value being significantly different to all comparative sample means closest to the archaic Petralona (0.08 eigenfactor units difference) (EAEHS z3.54, p0.004; EUEHS z3.35, p0.003; WAEHS z4.56, and NEAND Amud 1 (0.13) crania. Moreover, z-tests indicate that p0.005; NEAND z4.94, p0.002). for PC 1, the difference is significant between MLDG 1704 and Multivariate cranial comparisons. Table 11 summarises the H. sapiens mean (excluding Nazlet Khater 2) (z2.00, p0.03), but the results of principal components analysis of size-adjusted [27] not for NEAND (z-0.01) or ERECT (z-1.04). For PC 2, the variables for three sets of analyses comparing LL 1 or MLDG difference is significant between MLDG 1704 and NEAND (z2.19, 1704 to fossil specimens. The first analysis included LL 1 (Figure 8) p0.03), but not for H. sapiens (excluding Nazlet Khater 2: z0.42) or and 22 other crania, employed 9 variables (Table 11), and ERECT (z-0.03). generated three principal components. For principal component Comparison of crania to recent humans. In Tables 12–14 (PC) 1, the highest loading variables were frontal chord and frontal we compare cranial measurements of LL 1 and MLDG 1704 with arc, and these were contrasted mostly with measures of facial mixed sex recent humans: East Asian and Eskimo (Table 12), height (orbit height and upper facial height) and breadth (chiefly European and African (Table 13) and Australian (Table 14). These nasal breadth) (Table 11). For PC 2, facial breadth (orbit breadth samples are compiled from Howells’ [29] worldwide dataset (see and bimaxillary breadth) accounted for most of the variance also, Table 4). Overall, measures of facial width strongly (Table 11), while PC 3 contrasted maximum frontal breadth with distinguish LL 1 from recent humans, with its very broad face. bizygomatic breadth (Table 11). A bivariate plot of object scores Bizygomatic breadth (c144 mm) contrasts strongly with recent for PC 1 and PC 2 (Figure 8A) shows that PC 1 distinguishes human means (125–135 mm; z1.49–3.16, p0.06-0.0008). Bijugal crania belonging to H. sapiens from those of NEAND, H. breadth (c134 mm) is similarly enlarged (recent human means heidelbergensis sensu stricto, H. rhodesiensis and ERECT. Specimen 113–117 mm; z2.57–4.19, p0.005-0.00001). Its bimaxillary LL 1 lies within the range of H. sapiens, clustering with Skhul 5, breadth (c108 mm) contrasts strongly with recent sample means Mai Da Nuoc and Combe Capelle. A plot of PC 2 versus PC 3 (92–97 mm; z1.83–3.19, p0.03-0.0007). Bifrontal breadth

PLoS ONE | www.plosone.org 13 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 9. Facial skeleton measurements (mm) and indices (%) compared to Pleistocene modern humans (significant z-scores in bold)."

Measurement LL MLDG EAEHS EUEHS WAEHS AFEHS1 Abbrev. (Martin No.) 1 1704

1. Superior facial breadth [110] 109 11265(5) 11266(11) 12367(6) - SFB (M43) (106–119) (102–124) (96–110) - z-score/p LL 1 20.37/0.36 20.32/0.37 21.72/0.07 - z-score/p MLDG 1704 20.55/0.30 20.48/0.32 21.85/0.06 - 2. Bizygomatic breadth [144] - 13665(6) 14067(10) 148(4) 142(1) ZYB* (M45) (128–143) (129–156) (140–160) - z-score/p LL 1 1.48/0.09 0.54/0.29 - - 3. Postorbital constriction index (66) - 7363(6) 7364(10) 70(4) 79(1) (M9/M45) (70–78) (66–78) (66–76) - z-score/p LL 1 22.16/0.04 21.67/0.06 - - 4. Bimaxillary breadth [108] - 10566(7) 10168(8) 97(3) 100(1) ZMB* (M46) (95–114) (85–109) (90–110) - z-score/p LL 1 0.47/0.32 0.82/0.21 - 5. Upper versus mid-facial breadth 98 - 9367(5) 8865(5) 77(3) - index (M46/M43) (84–102) (80–92) (69–90) - z-score/p LL 1 0.65/0.27 1.83/0.07 - - 6. Superior facial height 64 - 6765(8) 6865(13) 7563(5) 79(1) SFH (M48) (61–75) (59–79) (72–79) - z-score/p LL 1 20.57/0.29 20.77/0.22 23.35/0.01 - 7. Facial index (44) - 5062(6) 4964(10) 51(4) 56(1) (M48/M45) (46–52) (44–58) (49–53) - z-score/p LL 1 22.78/0.01 21.19/0.13 - - 8. Orbit breadth 45 - 4462(8) 4364(14) 4562(5) - ORB (M51) (42–48) (38–48) (42–47) - z-score/p LL 1 0.47/0.32 0.48/0.31 n.a. - 9. Orbit height 34 - 3161(7) 3063(14) 3363(5) 34(1) OBH* (M52) (29–33) (26–36) (29–37) - z-score/p LL 1 2.81/0.01 1.29/0.11 0.30/0.38 - 10. Orbital index 76 - 7165(7) 7067(14) 7467(5) - (M52/M51) (67–79) (59–88) (65–84) - z-score/p LL 1 0.93/0.19 0.83/0.21 0.26/0.40 - 11. Maximum nasal width (32) - 2862(8) 2662(13) 3061(5) 29(1) NLB* (M54) (25–32) (22–30) (28–32) - z-score/p LL 1 1.89/0.05 2.89/0.006 1.83/0.07 - 12. Nasal height 45 - 5064(8) 5164(11) 4262(5) 56(1) NAH (M55) (46–58) (43–59) (42–47) - z-score/p LL 1 21.18/0.13 21.44/0.09 1.37/0.12 - 13. Nasal index (71) - 5664(8) 5066(11) 5663(5) - (M54/M55) (50–60) (44–63) (53–62) - z-score/p LL 1 3.54/0.004 3.35/0.003 4.56/0.005 -

"Fossil values in round brackets are estimates, values in square brackets estimated by measuring to the midline and doubling; above the line m6s(n), below the line (min.-max.); z-tests corrected for small comparative sample size; Bonferroni correction not employed as per [56]; sample abbreviations and compositions see Table4; data sources see Text S2. 1Mostly comprises measurements of Herto BOU-VP-16/1. *Equivalent to measurements of Howells [57] and employing his abbreviation. doi:10.1371/journal.pone.0031918.t009

(c106 mm) is greatly enlarged (96–100 mm; z2.00-1.50, p0.06- value being outside of the range of all samples. Finally, interorbital 0.02). Biorbital chord (c114 mm) also contrasts strongly with breadth in LL 1 (25 mm) is also broad compared with recent recent humans (97–99 mm; z3.40–5.66, p0.001-,0.00001), its populations (means 18–23 mm; z1.00–3.48, p0.15-0.0003).

PLoS ONE | www.plosone.org 14 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 10. Facial skeleton measurements (mm) and indices (%) compared to archaic hominins (significant z-scores in bold)."

Measurement LL MLDG NEAND EAMPH ERECT Abbrev. (Martin No.) 1 1704

1. Superior facial breadth [110] 109 11864{(27) - 123(2) SFB (M43) (107–128) - (121–125) z-score/p LL 1 21.96/0.03 -- z-score/p MLDG 1704 22.21/0.01 -- 2. Bizygomatic breadth [144] - 14568(6) 148(1) - ZYB* (M45) (130–153) - - z-score/p LL 1 20.12/0.45 - - 3. Postorbital constriction index (66) - 7462(6) 77(1) 71.5(2) (M9/M45) (71–77) - (69–74) z-score/p LL 1 23.70/0.006 -- 4. Bimaxillary breadth [108] - 11265{(13) - 107(2) ZMB* (M46) (104–120) - (98–116) z-score/p LL 1 20.77/0.22 - - 5. Upper versus mid-facial breadth 98 - 9565(10) - 87(2) index (M46/M43) (85–101) - (81–93) z-score/p LL 1 0.57/0.98 - - 6. Superior facial height 64 - 8765{(13) 74(1) 80.5(2) SFH (M48) (107–128) - (77–82) z-score/p LL 1 24.43/0.0004 -- 7. Facial index (44) - 5962(6) 50(1) 53(2) (M48/M45) (58–61) - (52–55) z-score/p LL 1 26.94/0.0004 -- 8. Orbit breadth 45 - 4463(12) 47(3) 42(2) ORB (M51) (40–49) (44–52) (40–44) z-score/p LL 1 0.32/0.37 - - 9. Orbit height 34 - 3861{(16) 36(3) 37(2) OBH* (M52) (36–41) (34–39) (34–40) z-score/p LL 1 23.88/0.0007 -- 10. Orbital index 76 - 8766(11) 71(2) 88(2) (M52/M51) (78–98) (67–76) (85–91) z-score/p LL 1 21.76/0.05 - - 11. Maximum nasal width (32) - 3264{(16) 31(1) 27(2) NLB* (M54) (23–39) - (24–30) z-score/p LL 1 n.a. - - 12. Nasal height 45 - 6163{(6) - 50.5(2) NAH (M55) (58–66) - (48–53) z-score/p LL 1 24.94/0.002 -- 13. Nasal index (71) - 5563(6) - 56.5(2) (M54/M55) (50–59) - (56–57) z-score/p LL 1 4.94/0.002 --

"Fossil values in round brackets are estimates, values in square brackets estimated by measuring to the midline and doubling; above the line m6s(n), below the line (min.-max.); z-tests corrected for small comparative sample size; Bonferroni correction not employed as per [56]; sample abbreviations and compositions see Table4; data sources see Text S2. *Equivalent to measurements of Howells [57] and employing his abbreviation. {T-test EUEHS and NEAND mean difference: one-tailed p,0.001. doi:10.1371/journal.pone.0031918.t010

Frontal chord length for LL 1 (112 mm) is long, but lies within chord to calculate the frontal curvature index (subtense/chord), it one standard deviation unit of all samples (means 108–111 mm; is clear that LL 1 exhibits an exaggerated degree of frontal z0.20–0.80). The value for frontal subtense is, however, high curvature (27%). This index distinguishes the Chinese fossil from (30 mm), and when this measurement is combined with frontal recent humans (means 23–25%; z1.00–2.00, p0.15–0.02). Glabella

PLoS ONE | www.plosone.org 15 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 11. Results of principal components analysis (two highest loading variables for each component in bold).

Component Component Component

123

LL 1 Set 1–9 variables x 23 objects Eigen score 4.07 1.41 1.13 % Variance 45.23 15.63 12.55 Variable loadings FRC 0.956 0.026 0.083 FAA 0.949 0.008 0.040 MFB 0.530 20.283 20.720 NPH 20.678 20.393 20.125 ORB 0.122 0.775 0.094 OBH 20.725 20.110 0.272 NLB 20.801 20.121 20.086 ZMB 20.449 0.646 20.093 ZYB 0.365 20.355 0.698 MLDG 1704 Set 1–8 variables x 23 objects Eigen score 3.61 2.44 1.11 % Variance 45.09 30.50 13.86 Variable loadings FRC 20.072 0.953 20.147 PAC 20.906 20.368 20.136 FAR 20.235 0.938 0.024 PAR 20.931 20.334 20.065 MFB 0.582 20.573 0.331 BPB 0.902 0.118 20.007 XFB 20.133 0.215 0.929 SFB 0.831 20.120 20.303 Set 2–6 variables x 43 objects Eigen score 2.77 1.84 - % Variance 46.21 30.68 - Variable loadings FRC 0.694 0.581 - PAC 20.890 0.019 - Figure 8. Object plots from principal components analysis FAA 0.496 0.825 - including LL 1 and 23 later Pleistocene fossil crania. (A) PC 1 versus 2, and (B) PC 2 versus PC 3 (NB: Gray star = LL 1; AFEHS = African PAA 20.878 0.258 - early H. sapiens; Amd = Amud; BG = Barma Grande; CC = Combe Capelle; MFB 0.494 20.782 - CM = Cro Magnon; Kei = Keilor; Hof = Hofmeyr; Kab = Kabwe; LaC = La XFB 0.486 20.378 - Chapelle; LaF = La Ferrassie; Liu = Liujiang; MDN = Mai Da Nuoc; Mld = Mladec; Oas = Oase; Pred = Predmost; S = Sangiran; SCr = Sima de doi:10.1371/journal.pone.0031918.t011 Los Huesos cranium; Skh = Skhul; Tab = Tabun; UC = Upper Cave- Zhoukoudian; and Zkd = Zhoukoudian ERECT). projection (4 mm) is well within the range of recent humans except doi:10.1371/journal.pone.0031918.g008 for Africans (non-African: means 3–5 mm, z1.00 to -1.00; African: 261 mm, z2.0, p0.02). Supraorbital projection (6 mm) is indistin- distinguishing the specimen from most recent human samples guishable from recent humans (means 5–7 mm; z0.00–1.00/ (means 47–51%; z-1.00 to -2.33, p0.16-0.01). Nasal height 21.00). Posterior breadth of the frontal bone (STB) is narrow (47 mm) is identical to the mean for Africans (4764 mm) and (103 mm) and closely resembles Australian (10267 mm; z0.14) similar to Australians (4863 mm), but distinct from all other and Eskimo (10167 mm; z0.28) means. samples (means 50–51 mm; z1.00 to -1.66). The nasal breadth of Facial height, or nasion-prosthion height in LL 1 (64 mm), is LL 1 (c32 mm) is broad and significantly different to all recent short, but well within the range of recent humans (means 62– human sample means (means 24–28 mm; z2.00–3.98, p0.02- 67 mm; z0.40 to -0.50). In contrast, its face is short relative to its 0.00006). The nasal index (breadth/height) (c71%) indicates the breadth (facial index = height/bizygomatic breadth) (c44%), nasal skeleton of LL 1 to be unusually broad and highly distinct

PLoS ONE | www.plosone.org 16 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

(3563 mm; z1.66, p0.04) means. Malar subtense, providing a measure of projection of the malar at its angle, is low in LL 1 (8 mm), in keeping with its broad and laterally flared zygomatics. Its value is significantly different to means for the East Asian (1262 mm; z-2.00, p0.02) and Eskimo (1262 mm; z-1.99, p0.02) samples. Specimen MLDG 1704 can be compared with mean values for five measurements (Tables 12–14). Bistephanic breadth (114 mm) lies within the range of recent humans (means 101–116 mm), but its value is significantly different to the Eskimo (10167 mm; z1.85, p0.03) and Australian (10267 mm; z1.71, p0.04) means. Maxi- mum frontal breadth (125 mm) is also wide in MLDG 1704 and distinguished from the means of recent humans (means 109– 115 mm; z1.00–3.73, p0.15-0.0008). Bifrontal breadth (107 mm) is characterised by statistically significant values when compared with all recent human samples (means 96–100 mm; z2.20 to -3.50, p0.01-0.0003). The frontal chord of MLDG 1704 is comparatively long (116 mm), but sits within the range of recent humans (means 108–110 mm; z1.00–1.60). The parietal chord is short (107 mm), but sits comfortably within the recent human range (110–114 mm; z-0.43 to -1.00). Finally, the ratio parietal/frontal chord in LL 1 is low (92%). Its value is distant from the means of most recent humans samples (non-Australian means 98–102 mm; z-1.00 to - 1.66, p0.15-0.05), being significantly different to the Australian mean (10566 mm; z-2.16, p0.01). Virtual endocast. A 3D virtual endocast was rendered from CT-scans of MLDG 1704 (Figure 11A–D; Figures S1, S2, S3, S4). Measurements of the frontal and parietal lobes were made and are compared in Table 15. They reinforce the visual impression of modern frontal lobes, which are long (86 mm), broad (121 mm) and tall (92 mm), being most like EUEHS (breadth: 12068 mm, z0.11; height: 10066 mm, z-0.15; chord length: 9067 mm, z- 0.52). Compared to recent East Asians, its frontal lobe is very long. The Maludong endocast is broader and taller than the Chinese Pleistocene H. sapiens cranium from Liujiang (breadth 115 mm, height 95 mm), and much broader than the endocast of the late Upper Pleistocene Japanese Minatogawa I cranium (112 mm). Its frontal lobe is, however, very distinct from the endocast of the H. rhodesiensis Kabwe cranium (breadth 108 mm, height 88 mm, Figure 9. Object plots from principal components analysis including MLDG 1704 and 23 later Pleistocene fossil crania. (A) chord 78 mm), broader and longer than NEAND endocasts PC 1 versus 2, and (B) PC 2 versus PC 3 (NB: Gray star = MLDG 1704; (breadth 107 mm, chord 82 mm) and broader, longer and taller BG = Barma Grande; CM = Cro Magnon; GdE = Grotte de Enfants; than ERECT (breadth: 9968 mm; z2.64, p0.01; height: LaC = La Chapelle; LaF = La Ferrassie; Liu = Liujiang; LQ = La Quina 5; 74611 mm; 2.18, p0.02; chord: 7666 mm; z1.60). MC = Monte Circeo; MDN = Mai Da Nuoc; Nea = Neandertal; Pet = Pe- In contrast, the parietal lobes of MLDG 1704 are very short tralona; Pred = Predmost; S = Sangiran; Skh = Skhul; UC = Upper Cave- (99 mm), contrasting with the long parietal lobes of EUEHS Zhoukoudian; and Zkd = Zhoukoudian ERECT). doi:10.1371/journal.pone.0031918.g009 (12267 mm; z-3.00, p0.01), recent Chinese (10664 mm; z-1.72, p0.04) and recent Japanese (10767 mm; z-1.13). The parietal chord length for NEAND is also moderate (106 mm), like Liujiang from recent humans (means 45–59%; z2.40–6.47, p0.008- (107 mm) and Minatogawa 1 (103 mm). While the parietal lobes , 0.00001). The value for orbit height (34 mm) sits well within of MLDG 1704 are short, much shorter even than Kabwe the range of recent humans (32–36 mm; z0.00 to -1.00), while (104 mm), they are longer than ERECT (8767 mm, z1.64). orbit breadth (44 mm) contrasts strongly with them (means 39– Figure 11E is a bivariate plot of the breadth of the frontal lobes 41 mm; z1.49–2.50, p0.06-0.006). Orbit index (height/breadth) versus frontal height. It confirms the modern size and shape of the reinforces the relatively short (for its breadth) left orbit of LL 1 frontal of MLDG 1704, its value sitting well within the range of (76%), its value being significantly different to the sample means EUEHS (i.e. Predmost crania) and recent Japanese. Figure 11F for East Asian (8765%; z-2.20, p0.01) and Eskimo (8765; z-2.19, compares frontal chord and parietal chord dimensions of the p0.01). The simotic chord is small in LL 1 (4 mm), emphasising the endocast and shows MLDG 1704 to be just within the range of narrowness of its nasal bones superiorly. Its value strongly recent Chinese, outside of the range of fossil H. sapiens, and very distinguishing the specimen from East Asian (862 mm; z-2.00, close to the ERECT specimen Zhoukoudian 10. p0.02), Australian (962 mm; z-2.50, p0.006), European (962 mm; Mandibles. Table 16 compares a range of commonly z-2.50, p0.006) and African (963 mm; z-1.66, p0.04) samples. employed mandibular characters and Table 17 body metrics for The inferior length of the zygomatic bone of LL 1 (26 mm) is distinguishing among later Pleistocene hominins. While the short and distinguishes the fossil from East Asian (3564 mm; symphyseal region of LL 1 has been damaged, in our z1.67, p0.04), Eskimo (4064 mm, z-3.48, p0.0003) and European judgement, it would probably have possessed a chin of Rank 3

PLoS ONE | www.plosone.org 17 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Figure 10. Object plot from principal components analysis including MLDG 1704 and 43 later Pleistocene fossil crania. NB: Gray star = MLDG 1704; AFEHS = African early H. sapiens; Amd = Amud; BG = Barma Grande; Buk = Buku; CC = Combe Capelle; CM = Cro Magnon; Kei = Keilor; Hof = Hofmeyr; GdE = Grotte de Enfants; Kab = Kabwe; LaC = La Chapelle; LaF = La Ferrassie; Liu = Liujiang; LQ = La Quina 5; MC = Monte Circeo; MDN = Mai Da Nuoc; Mld = Mladec; Nea = Neandertal; Ng = Ngandong; NK = Nazlet Khater 2; Oas = Oase; Pet = Petralona; Pred = Predmost; S = Sangiran; SCr = Sima de Los Huesos cranium; Skh = Skhul; Sp = Spy; Tab = Tabun; UC = Upper Cave-Zhoukoudian; and Zkd = Zhoukoudian ERECT. doi:10.1371/journal.pone.0031918.g010

[30]. The chin of MLDG 1706 is Rank 3 [30], and while relatively scarred in both Maludong mandibles, they lack a prominent common among Eurasian early H. sapiens (29.2–49.5%), the superior pterygoid tubercle (present: NEAND 81.2%, Eurasian Chinese Tianyaun 1 and Zhirendong 3 mandibles possess Rank 4 early H. sapiens 10–76.7%). Finally, the crest of the mandibular chins. Specimens LL 1 and MLDG 1706 lack a vertical keel and notch meets the condyle laterally in MLDG 1679, but it is more lateral tubercles, features which form the major components of the medially located in MLDG 1706. Medial placement of the crest is modern human ‘inverted-T’ form chin [31]. In inferior view, the found frequently in NEAND (63% presence, versus 100% absence anterior border of the body (beneath the symphysis) is rounded in in Western and European H. sapiens) and characterises the Dar-es- both LL 1 and MLDG 1706, more like the condition seen in Soltane 5 mandible with its apparent archaic affinities [1,32]. archaic hominins [31]. The mental foramen is located below P4/ Internally, the alveolar plane of LL 1 and MLDG 1706 is M1 in LL 1 and MLDG 1706. Location of this foramen mesial to posteriorly inclined and the transverse tori are thickened. This is a M1 is characteristic of early H. sapiens (88–100% presence versus common feature among archaic later Pleistocene hominins such as 12.2% NEAND). Mandibular foramen bridging is absent in Te´mara 1 (North Africa), but is largely absent from early H. sapiens MLDG 1679, but present in MLDG 1706 (cannot be scored on [32]. Externally, the symphysis is somewhat undercut in lateral LL 1). Absence of bridging is common in NEAND (57.1% aspect, and its anterior symphyseal angle is low (77u), a value closest presence), but rare in Eurasian early H. sapiens (0–7% presence). to NEAND (80.867.3u; z-0.51) and the Te´mara 1 mandible (80u). Both Maludong mandibles show asymmetry of the mandibular In contrast, Pleistocene H. sapiens angles are more acute (means notch. However, the coronoid process of MLDG 1679 is 86.6–96.5u; z-1.34 to -3.02), as seen also in the East Asian mandibles disproportionately large, a feature common among NEAND, Tianyuan 1 (,96u) and Zhirendong 3 (91u). Body height (26.9 mm) while in MLDG 1706 it is greatly reduced, the H. sapiens condition. and thickness (13.3 mm) at the level of the mental foramen in In LL 1, the coronoid process is large, but its proportions cannot MLDG 1706 are comparatively low, showing the specimen to be be assessed owing to an absence of the notch and condylar process. similar to modern humans in its size. Its body height sits just outside Specimens LL 1 and MLDG 1679 possess a retromolar space (M3 of the range of Pleistocene East Asian H. sapiens mandibles (range is uncovered [21]), a common characteristic of NEAND (presence: 27.4–33.7 mm), but body thickness is comfortably within their 75%, versus 32.9–40% in early H. sapiens). In MLDG 1706, the range (11.3–14.4 mm). Body height (28 mm) and thickness (14 mm) M3 is partially covered; scored here as absence of a retromolar measured slightly posterior to the mental foramen in LL 1 is similar space. While the medial pterygoid attachment area is strongly to the Maludong specimen (Table 16).

PLoS ONE | www.plosone.org 18 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 12. Comparison of LL 1 and MLDG 1704 with Howells [29] recent East Asian and Eskimo samples (significant z-scores in bold)."

Howells Fossil E. Asian Eskimo

Measurement (Abbrev.) Value (n = 681) (n = 108) m6s zp m6s zp

LL 1 Bistephanic breadth (STB) 103 11267 21.28 0.09 10167 0.28 0.38 Bizygomatic breadth (ZYB) [144] 13268 1.50 0.06 13566 1.49 0.06 Nasion-prosthion height (NPH) 64 6766 20.50 0.30 6964 21.24 0.10 Facial index (NPH/ZYB) 44 5163 22.33 0.01 5163 22.32 0.01 Nasal height (NLH) 47 5164 21.50 0.06 5263 21.66 0.05 Bijugal breadth [JUB] [134] 11667 2.57 0.005 11766 2.82 0.002 Orbit height (OBH) 34 3462 n.a. n.a. 3662 21.00 0.16 Orbit breadth (OBB) 44 3962 2.50 0.006 4162 1.49 0.06 Orbit index (OBH/OBB) 76 8765 22.20 0.01 8765 22.19 0.01 Nasal breadth (NLB) [32] 2762 2.50 0.006 2462 3.98 0.00006 Nasal index (NLB/NLH) 71 5265 3.80 0.00005 4564 6.47 ,0.00001 Bimaxillary breadth (ZMB) [108] 9766 1.83 0.03 9666 1.99 0.02 Bifrontal breadth (FMB) [106] 9665 2.00 0.02 9764 2.24 0.01 Biorbital breadth (EKB) [114] 9765 3.40 0.0003 9864 3.98 0.00006 Interorbital breadth (DKB) 25 2162 2.00 0.02 1862 3.48 0.0003 Simotic chord (WNB) 4 862 22.00 0.02 662 21.00 0.16 Malar length, inferior (IML) (26) 3564 1.67 0.04 4064 23.48 0.0003 Supraorbital projection (SOS) 6 661 n.a. n.a. 561 1.00 0.16 Malar subtense (MLS) 8 1262 22.00 0.02 1262 21.99 0.02 Glabella projection (GLS) 4 361 1.00 0.15 361 1.00 0.16 Frontal chord (FRC) 112 11065 0.60 0.27 11165 0.20 0.42 Frontal subtense (FRS) 30 2663 1.33 0.09 2763 1.00 0.15 Frontal curvature index (FRS/FRC) 27 2362 2.00 0.02 2462 1.49 0.06 MLDG 1704 Bistephanic breadth (STB) 114 11267 0.29 0.38 10167 1.85 0.03 Maximum frontal breadth (XFB) 125 11567 1.43 0.07 11064 3.73 0.0001 Bifrontal breadth (FMB) 107 9665 2.20 0.01 9764 2.49 0.007 Frontal chord (FRC) 116 11065 1.40 0.08 11165 1.00 0.16 Parietal chord (PAC) 107 11067 20.43 0.33 11366 21.00 0.16 Parietal/frontal chord index (PAC/FRC) 92 10166 21.50 0.06 10266 21.66 0.05

"Fossil values in round brackets are estimates, values in square brackets estimated by measuring to the midline and doubling; sample compositions see Table 4; descriptive statistics for Howells’ samples calculated by us from raw data; z-test results do not employ Bonferroni correction as per [56]. doi:10.1371/journal.pone.0031918.t012

Dentition. The mostly well preserved, but worn, dental z-0.56) and a Middle Palaeolithic H. sapiens sample (8.360.8 mm; crowns of LL 1 and MLDG 1706 (Figure 6–7), and an isolated z-0.36). In contrast, its P3 BL diameters (9.3/9.4 mm) are large like maxillary third molar (MLDG 1747: Figure 12), also reveal NEAND (9.060.7; z0.56) and ERECT (10.060.6 mm; z-0.97). Its important information about their morphology and affinities. P3 crown width is significantly different to mean values for EAEHS Buccolingual (BL) crown diameters and descriptive statistics for (8.460.1 mm; z-9.13, p0.0003) and a Western Middle Upper comparative samples are provided in Table 18 (mandibular Palaeolithic Human sample (8.560.5; z-1.75, p0.04). The M2 dentition) and Table 19 (maxillary dentition). crown of MLDG 1679 is broad (11.9 mm), but its value is not The LL 1 I2 crown (6.7 mm) is narrow, its value sitting well significantly different to comparative sample means (11.0– within the range of comparative samples except NEAND with 12.8 mm; although it is approaching significance for EAEHS their broad incisors (H. sapiens means 6.9–7.2 mm; z-0.20 to -0.95; z1.87, p0.05). Mandibular M3 crown BL diameters for LL 1 (10.7/ NEAND 7.760.5; z-1.97, p0.02; ERECT 7.160.5 mm; z-0.77). 10.3 mm) and MLDG 1679 (11.6 mm) are moderate to large. The The mandibular canine crowns of LL 1 are also small (8.4/ crown of the former specimen is not significantly larger than 7.6 mm), and while not significantly different to any comparative comparative means (10.4–11.5 mm; z-0.23 to -0.88), while the sample mean, its BL diameter is closest to EAEHS (8.360.5 mm; latter is distinct from EAEHS (10.460.4 mm; z2.81, p0.01).

PLoS ONE | www.plosone.org 19 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 13. Comparison of LL 1 and MLDG 1704 with Howells [29] recent European and African samples (significant z-scores in bold)."

Howells Fossil European African

Measurement (Abbrev.) Value (n = 317) (n = 365) m6s zp m6s zp

LL 1 Bistephanic breadth (STB) 103 11667 21.85 0.03 10867 0.71 0.23 Bizygomatic breadth (ZYB) [144] 13066 2.33 0.01 12566 3.16 0.0008 Nasion-prosthion height (NPH) 64 6665 20.40 0.34 6266 0.33 0.36 Facial index (NPH/ZYB) 44 5163 22.33 0.01 5064 21.50 0.06 Nasal height (NLH) 47 5063 1.00 0.15 4764 n.a. n.a. Bijugal breadth [JUB] [134] 11465 3.99 0.00004 11365 4.19 0.00001 Orbit height (OBH) 34 3362 0.50 0.30 3362 0.50 0.30 Orbit breadth (OBB) 44 3962 2.50 0.006 3962 2.50 0.006 Orbit index (OBH/OBB) 76 6669 1.11 0.13 8466 21.33 0.09 Nasal breadth (NLB) [32] 2562 3.49 0.0002 2862 2.00 0.02 Nasal index (NLB/NLH) 71 5064 5.24 ,0.00001 5965 2.40 0.008 Bimaxillary breadth (ZMB) [108] 9265 3.19 0.0007 9465 2.80 0.002 Bifrontal breadth (FMB) [106] 9764 2.25 0.01 9864 2.00 0.02 Biorbital breadth (EKB) [114] 9763 5.66 ,0.00001 9764 4.24 0.0001 Interorbital breadth (DKB) 25 2262 1.50 0.06 2362 1.00 0.15 Simotic chord (WNB) 4 962 22.50 0.006 963 21.66 0.04 Malar length, inferior (IML) (26) 3563 1.66 0.04 3664 1.00 0.15 Supraorbital projection (SOS) 6 661 n.a. n.a. 661 n.a. n.a. Malar subtense (MLS) 8 1062 1.00 0.15 1162 21.50 0.06 Glabella projection (GLS) 4 361 1.00 0.15 261 2.00 0.02 Frontal chord (FRC) 112 11065 0.40 0.34 10865 0.80 0.21 Frontal subtense (FRS) 30 2663 1.33 0.09 2763 1.00 0.15 Frontal curvature index (FRS/FRC) 27 2362 2.00 0.02 2562 1.00 0.15 MLDG 1704 Bistephanic breadth (STB) 114 11667 20.29 0.38 10867 0.89 0.19 Maximum frontal breadth (XFB) 125 11966 1.00 0.15 11166 2.33 0.01 Bifrontal breadth (FMB) 107 9764 2.50 0.006 9864 2.25 0.01 Frontal chord (FRC) 116 11065 1.20 0.11 10865 1.60 0.05 Parietal chord (PAC) 107 11166 20.67 0.25 11166 20.67 0.25 Parietal/frontal chord index (PAC/FRC) 92 10166 21.50 0.06 9866 21.00 0.15

"Fossil values in round brackets are estimates, values in square brackets estimated by measuring to the midline and doubling; sample compositions see Table 4; descriptive statistics for Howells’ samples calculated by us from raw data; z-test results do not employ Bonferroni correction as per [56]. doi:10.1371/journal.pone.0031918.t013

The measurable maxillary crowns of LL 1 are comparatively also taurodont (Figure 12), its three roots being fused for most of broad. Its P4 BL (11.0 mm) is most like ERECT (11.561.0 mm; z- their course. Taurodontism is rare among recent and EUEHS 0.49, p0.31) and is distinct from H. sapiens (Qafzeh-Skhul humans [34–35], but is commonly considered a distinguishing 10.260.8 mm; z0.95; Upper Palaeolithic H. sapiens 9.960.6 mm; feature of NEAND [35–36]. z1.80, p0.04) and NEAND (10.060.7 mm; z1.40). In contrast, its 1 M crown is narrow (11.7 mm), and while its value is well within the Discussion range of all comparative samples, it is closest to the NEAND mean (12.060.8; z-037). The BL diameter of an isolated M3 MLDG 1747 The partial human skull from Longlin Cave and the human (12.5 mm) is comparatively large, but sits within the range of all calotte, partial mandibles and teeth from Maludong both present a samples listed in Table 19, being equally close to the Qafzeh-Skhul range of individual features and a composite of characters not seen (11.760.6; z0.42) and NEAND (11.961.4; z0.42) means. among Pleistocene or recent populations of H. sapiens. It is clear Measurements made on CT-scans of the in situ M3 of MLDG that they share no particular affinity with either Pleistocene East 1679 (not given) indicate that this tooth is taurodont (Taurodontism Asians, such as Liujiang or Upper Cave 101, or recent East Asians. index [33] 26.1%, or hypotaurodont). Additionally, MLDG 1747 is These features belong to multiple developmental-functional

PLoS ONE | www.plosone.org 20 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 14. Comparison of LL 1 and MLDG 1704 with Howells variance), shows LL 1 and MLDG 1704 to be at the edge of [29] recent Australian samples (significant z-scores in bold)." variation within Pleistocene H. sapiens, and in some analyses, on the edge also of H. erectus variability. A weaker phenetic signal, revealed particularly by principal component 3 (,12–14% of total Howells Fossil Australian variance), shows them to exhibit a unique cranial shape among all later Pleistocene hominins. Measurement (Abbrev.) Value (n = 165) A range of features support the conclusion that these remains m6s zp show affinities to H. sapiens: LL 1 N Neurocranium: moderately projecting and laterally thin Bistephanic breadth (STB) 103 10267 0.14 0.44 supraorbital part, which has the bipartite form in MLDG Bizygomatic breadth (ZYB) [144] 13167 1.85 0.03 1704; frontal bone with a moderate chord and arc length, but Nasion-prosthion height (NPH) 64 6265 0.40 0.34 broad maximum width; and an endocast with long, broad and tall frontal lobes. Facial index (NPH/ZYB) 44 4763 21.00 0.16 N Viscerocranium: narrow superior facial breadth; vertically Nasal height (NLH) 47 4863 20.33 0.37 short facial skeleton (superior facial height, orbit height and Bijugal breadth [JUB] [134] 11666 2.99 0.001 nasal height); and moderate nasal breadth relative to height. Orbit height (OBH) 34 3262 1.00 0.16 N Mandible: mesial position of the mental foramen; and Orbit breadth (OBB) 44 4162 1.50 0.06 absence of a medial pterygoid tubercle. Orbit index (OBH/OBB) 76 8066 20.66 0.25 N Dentiton: small (narrow) anterior dental crowns. Nasal breadth (NLB) [32] 2762 2.49 0.006 At the same time, the Longlin and Maludong fossils possess Nasal index (NLB/NLH) 71 5865 2.59 0.005 many features that are either rare or absent among Pleistocene Bimaxillary breadth (ZMB) [108] 9466 2.33 0.01 and recent H. sapiens, many of them being putative plesiomorphies Bifrontal breadth (FMB) [106] 10064 1.50 0.06 of later Homo. These include: Biorbital breadth (EKB) [114] 9964 3.74 0.0001 N Neurocranium: moderate endocranial volume; highly Interorbital breadth (DKB) 25 2162 1.99 0.02 arched frontal squama; short parietal bones; endocast with Simotic chord (WNB) 4 962 22.50 0.006 short parietal lobes; narrow postorbital region; and absence of Malar length, inferior (IML) (26) 3964 0.25 0.40 a bipartite supraorbital morphology in LL 1. Supraorbital projection (SOS) 6 761 21.00 0.16 N Cranial base (LL 1 only): a mandibular fossa that is long (A- Malar subtense (MLS) 8 1162 21.50 0.06 P), broad (M-L) and deep (S-I). Glabella projection (GLS) 4 561 21.00 0.16 N Viscerocranium (LL 1 only): strong alveolar prognathism; Frontal chord (FRC) 112 10965 0.60 0.27 flat mid-face, both at the nasal root and aperture and zygomatic process of the maxilla; broad facial skeleton Frontal subtense (FRS) 30 2562 2.50 0.006 (interorbital, bizygomatic and bimaxillary); very narrow nasal Frontal curvature index (FRS/FRC) 27 2362 1.99 0.02 bones; broad piriform aperture; absence of a canine fossa, and MLDG 1704 possessing a deep sulcus maxillaris; zygomatic arch is laterally Bistephanic breadth (STB) 114 10267 1.71 0.04 flared; zygomatic strongly angled such that its inferior margin Maximum frontal breadth (XFB) 125 10965 3.19 0.0008 sits well lateral to the superior part; zygomatic tubercle is small Bifrontal breadth (FMB) 107 10064 23.50 0.0003 and sits lateral to a line project from the orbital pillar (anterior aspect); the anterior masseter attachment area is marked by a Frontal chord (FRC) 116 10965 1.40 0.08 broad and deep sulcus; strong transverse incurvation of lateral Parietal chord (PAC) 107 11466 21.16 0.12 orbital pillar (lateral aspect); and anterior placement of the Parietal/frontal chord index (PAC/FRC) 92 10566 22.16 0.01 anterior wall of the zygomaticoalveolar root (above P4/M1). N Mandible: absence of a sagittal keel and distinct lateral "Fossil values in round brackets are estimates, values in square brackets estimated by measuring to the midline and doubling; sample compositions see tubercles; small chin (MLDG 1706 Rank 3, LL 1 Rank ?3); Table 4; descriptive statistics for Howells’ samples calculated by us from raw mandibular foramen bridging (MLDG 1706); thickened data; z-test results do not employ Bonferroni correction as per [56]. transverse tori; asymmetrical mandibular notch (MLDG doi:10.1371/journal.pone.0031918.t014 1679); retromolar space; crest of the mandibular notch positioned laterally (MLDG 1679); and a low anterior complexes [37], spanning the neurocranial vault, including symphyseal angle (MLDG 1706). endocranium, cranial base, facial skeleton, mandible and denti- N Dentition: broad post-canine crowns (large BL diameters); tion. Where they can be assessed, metrical dimensions involved are and taurodont molars. characterised mostly by moderate to high heritability [38–41]. Given their morphological similarity, close geographical proximity The finding of human remains with such a combination of (,300 km apart) and young geological age (i.e., Pleistocene- modern (H. sapiens) and archaic (putative plesiomorphic) characters Holocene transition), it seems likely that both samples belong to is unusual, especially in Eurasia. In Africa, there are several the same population. Pleistocene remains that also combine modern features with Multivariate analysis of vault shape, a method shown to track putative later Homo plesiomorphies: from Klasies River Mouth neutral genetic distances [42], indicates a somewhat mixed picture Cave [43–44] and Hofmeyr [45] (), Iwo Eleru () with respect to the phenetic affinities of LL 1 and MLDG 1704. [46], Nazlet Khater (Egypt), and Dar-es-Soltane and Te´mara The dominant phenetic signal in these analyses, as indicated by () [1,28,45]. Most of them are, however, much older than the first principal component (accounting for 45–46% of total Longlin and Maludong: Dar-es-Soltane and Te´mara are undated,

PLoS ONE | www.plosone.org 21 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Figure 11. Virtual endocast of MLDG 1704. Left panel: (A) left lateral aspect, (B) superior aspect, (C) anterior aspect, and (D) posterior aspect. Right panel: (E) plot of frontal breadth versus frontal height, and (F) plot of frontal chord versus parietal chord (Gray star = MLDG 1704; ellipses are ranges for samples; Chinese = recent Chinese; Japanese = recent Japanese; Kab = Kabwe; Liu = Liujiang; Min 1 = Minatogawa 1; Nea = Neandertal Pred = Predmost; Zkd = Zhoukoudian). doi:10.1371/journal.pone.0031918.g011 but they were associated with the Aterian lithic assemblage which 28]. Some of them, such as from Skhul and Qafzeh () and has recently been dated to between 10763kaand9664kaat Pestera cu Oase () have been included in our analyses, another Moroccan site (La Grotte des Contrebandiers) [47]; the and overall seem to be metrically well within the range of Klasies River Mouth remains are from two units dating .101 ka Pleistocene H. sapiens (e.g. Figures 8–9). The former (Levantine) and .64–104 ka [48]; Nazlet Khater 2 is perhaps <42 ka [28]; and samples do, however, show some similarities to LL 1 and MLDG Hofmeyr 36.263.3 ka [49]. However, the recently described Iwo 1704 in univariate comparisons. Eleru calvaria has been dated ,16.3-11.7 ka [46] and is clearly of How might the presence of this unusual morphology during the similar age to the Chinese remains. Pleistocene-Holocene transition of East Asia be explained? The Various Upper Pleistocene fossils outside of Africa have also remains from Longlin and Maludong could represent very robust been described as exhibiting an unusual mosaic of characters [e.g. individuals within a previously unknown Epipalaeolithic popula-

PLoS ONE | www.plosone.org 22 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 15. Endocast chord measurements (mm) compared developmental-functional complexes (as noted above). Moreover, (significant z-scores in bold)." this hypothesis could also be invoked to explain the morphology of remains from Klasies River Mouth Cave, Hofmeyr, Iwo Eleru, Nazlet Khater, Dar-es-Soltane, Te´mara and Zhirendong, but has not because many of their archaic features are rare or absent Sample Frontal Frontal Frontal Parietal among H. sapiens. The same situation applies to the Longlin and Breadth Height Chord Chord Maludong remains, as shown strongly here. MLDG 1704 121 99 86 99 In our opinion, there are more plausible explanations. One Liujiang 115 95 89 107 possibility is that the Longlin and Maludong remains represent a late surviving archaic population, perhaps similar to that sampled Minatogawa 1 112 - 71 103 at Dar-es-Soltane and Te´mara [1,28,32]. Unfortunately, little is Kabwe 108 88 78 104 known of the morphology of these North African remains, and EUEHS 12068(5) 10066(5) 9067(5) 12267(5) their affinities and taxonomy are unclear [1,28,32]. Within East (106–126) (93–109) (84–101) (105–122) Asia, the recently described mandibular fragment from Zhiren- z-score/p 0.11/0.45 20.15/0.44 20.52/0.31 23.00/0.01 dong also possesses a mosaic of modern and plesiomorphic characters making its taxonomic status problematic [3,10–11]. It Recent Chinese 11265(31) 9264(31) 7965(31) 10664(31) has, although, been dated on stratigraphic grounds to .100 ka (103–122) (86–100) (69–89) (98–114) [10], similar in age to the North African Aterian assemblage, but z-score/p 1.77/0.04 1.72/0.04 1.38/0.08 21.72/0.04 much older than Longlin and Maludong. Another recently Recent Japanese 12064(32) 9565(32) 7765(32) 10767(32) described specimen from the site of Salkhit () has also (112–129) (83–105) (64–84) (95–119) been described as belonging to an unspecified archaic taxon [50]. , z-score/p 0.25/0.40 0.79/0.21 1.77/0.04 21.13/0.13 Dating is uncertain, although, a preliminary date of 20 ka has * apparently been reported [3]. Moreover, doubts about its archaic NEAND 107(6) - 82(6) 106(3) affinities have been expressed [3] (Note: we have been unable to * (103–111) - (72–92) (104–110) include this specimen in our analyses as we found errors in the ERECT 9968(12) 74611(12) 7666(11) 8767(11) measurements of this and other specimens included in Table 1 of (84–110) (44–88) (68–85) (70–96) Coppens et al. [50]). z-score/p 2.64/0.01 2.18/0.02 1.60/0.07 1.64/0.06 Another possible explanation is that the unusual morphology of the Longlin and Maludong remains results from the retention of a large "Above the line m6s (n), below the line (min.-max.); z-test results do not number of ancestral polymorphisms in a population of H. sapiens.The employ Bonferroni correction as per [56]; data sources see Text S2. concept of incomplete lineage sorting is commonly invoked to explain *Data not available because published height for this taxon has been measured from the frontal pole-occipital pole; cannot be measured in MLDG 1704 owing morphologically mixed groups where the features of interest are to absence of the occipital bone. present also in allopatric populations belonging to the same taxon doi:10.1371/journal.pone.0031918.t015 [51]. Related to this, recent morphological studies have suggested that Pleistocene H. sapiens was deeply geographically subdivided tion in southwest China. We consider this to be an unsatisfactory within Africa prior to its dispersal into Eurasia [52]. This explanation explanation because of the presence of several apparently unique has also been invoked to explain the unusual morphology of the Iwo features combined with an unusual mixture of modern and archaic Eleru calvaria [46]. The morphology documented at Longlin and features is seen in several specimens and spans multiple Maludong might be interpreted as consistent with this hypothesis, the

Table 16. Mandibular body traits compared."

Sample Mentum Mental foramen Mandibular Mandibular notch Retromolar Medial pterygoid osseum rank location foramen bridging symmetrical space tubercle

(% rank 4) (% mesial of M1) (% absent) (% present) (% absent) (% absent)

LL 1 ?3 P4/M1 - - ?Present - MLDG 1679 - - Absent Asymmetrical Present Absent

MLDG 1706 3 P4/M1 Present Asymmetrical Absent Absent

Tianyuan 1 4 P4/M1 Absent Symmetrical - Absent

Zhirendong 3 4 P4 ---- EAEHS 50.5(4) 90.9(11) 100(3) 100(3) - 33.3(3) EUEHS 70.8(4) 92.0(25) 100(16) 100(17) 77.1(17) 90.0(10) Western EHS 85.7(7) 100(5) 83.3(3) 66.7(3) 60(5) 100(3) AFEHS 68.8(8) 87.7(7) 100(5) 100(4) - 100(6) NEAND 0.0(23) 12.2(31) 42.9(21) 30.8(13) 25(28) 18.8(16) ERECTˆ - 33(12) - - - -

"Sample abbreviations and compositions see Table 4, except Western EHS ( = MIS3 early modern humans [9–10]); data sources [9–10] and see Text S2. ˆMost ERECT mandibles possess multiple mental foramina. We have scored a mandible as having a mental foramen mesial to M1 only when all foramina are in this position. doi:10.1371/journal.pone.0031918.t016

PLoS ONE | www.plosone.org 23 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 17. Mandibular body measurements compared (significant z-scores in bold)."

Sample Anterior symphyseal Body height at Body thickness at angle mental foramen mental foramen

(6) (mm) (mm)

LL 1 - (28)* 14* MLDG 1706 ,77 26.9 13.3 Tianyuan 1 ,96 28.7 11.3 Zhirendong 3 91 27.4 16.0 EAEHS 94 29.0,31.0,33.7 12.0,13.0,14.4 EUEHS 96.566.2(12) 31.664.4(12) 12.461.4(11) Z-score/p LL 1 - 20.79/0.22 1.09/0.14 Z-score/p MLDG 1706 23.02/0.005 - 0.62/0.27 Western EHS 89,91 27.5,33.2,35.3 11.6,12.2,15.7 AFEHS 86.466.4(5) 35.0,36.0,40.5 13.2,15.0,16.6 Z-score/p MLDG 1706 21.34/0.12 - - NEAND 80.867.31(18) 32.363.6(26) 15.561.81(26) Z-score/p LL 1 - 21.17/0.12 20.82/0.21 Z-score/p MLDG 1706 20.51/0.30 - 21.20/0.12 ERECT 68.9612.7(6) - - Z-score/p MLDG 1706 0.59/0.29 - -

"Sample abbreviations and compositions see Table 4, except Western EHS ( = MIS3 early modern humans [9–10]); data sources [9–10] and see Text S2; m6s(n); z-test results do not employ Bonferroni correction as per [56]. *Value taken slightly distal to mental foramen owing to damage. 1T-test EUEHS and NEAND mean difference: one-tailed p,0.05-0.01. doi:10.1371/journal.pone.0031918.t017

Chinese remains perhaps sampling a previously unknown human population (or migration?) that may not have contributed genetically to recent East Asians. Ancient DNA could allow for a test of this idea, however, our ongoing attempts to extract DNA from a specimen from Maludong have so far proven unsuccessful owing to a lack of recoverable genetic material. Either way, the presence of the unusual morphology sampled at Longlin and Maludong during the Pleistocene-Holocene transition indicates that the evolutionary history of humans in East Asia is more complex than has been understood until now. It further highlights the need for much more research in the region as a matter of priority.

Methods Radiocarbon dating Fifteen charcoal samples for AMS radiocarbon assay were prepared and measured at the ANTARES-STAR Accelerator Mass Spectrom- etry Facility at the Australian Nuclear Science and Technology Organisation described in [60]. All samples were pre-treated and converted to graphite following methods described by [61]. The external surfaces of charcoal pieces selected for assay were scraped with a cleaned scalpel to remove sediment and soil attached to charcoals. The samples were then cut into smaller pieces to increase surface areas for more efficient chemical pre-treatment. Each sample was then treated with an acid-base-acid sequence as follows:

N 2 M HCl at 60uC for 2 hours to remove carbonate and any infiltrated fulvic acid contaminants, Figure 12. Isolated M3 –specimenMLDG1747(scale N 0.5–4% NaOH at 60uC for 10 hours to remove infiltrated bar = 1 cm) exhibiting marked taurodontism. fulvic and humic acid contaminants. This treatment is doi:10.1371/journal.pone.0031918.g012 commenced with a very weak alkali solution of 0.5% NaOH

PLoS ONE | www.plosone.org 24 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 18. Comparison of mandibular dental crown buccolingual diameters (mm) (significant z-scores in bold)."

Fossil/Sample I2 C P3 M2 M3

LL 1 6.7 8.4/7.61 9.3/9.51 - 10.7/10.31 MLDG 1679 - - - 11.9 11.6 Tianyuan 1 7.1 8.9 8.2 10.7 11.3 Zhirendong 1, 2 - - - 10.3 10.1, 0.3 Eastern EMH 6.960.3{(6) 8.360.5{(6) 8.460.1{(5) 11.160.4{(7) 10.460.4{(7) z-score/p LL 1 20.64/0.28 20.56/0.30 29.13/0.0003 - 20.23/0.41 z-score/p MLDG 1679 - - - 1.87/0.05 2.81/0.01 Western MUP{ 6.860.5(22) 8.660.7(19) 8.560.5(18) 11.060.8(28) 10.860.9(17) z-score/p LL 1 20.20/0.42 20.08/0.46 21.75/0.04 - 20.88/0.19 z-score/p MLDG 1679 - - - 1.11/0.13 0.86/0.20 Western EMH{ 7.260.5(5) 8.760.8(6) 8.4,9.0 11.261.1(7) 10.0,11.7,14.2 z-score/p LL 1 20.91/0.20 20.81/0.22 - - - z-score/p MLDG 1679 - - - 0.60/0.28 - Middle Palaeolithic MH{ 7.260.5(10) 8.360.8(10) 8.860.5(8) 11.060.7(10) 10.860.6(8) z-score/p LL 1 20.9/0.18 20.36/0.36 1.13/0.14 - 20.88/0.19 z-score/p MLDG 1679 - - - 1.23/0.12 1.26/0.12 Neandertals{ 7.760.5(28) 8.960.7(33) 9.060.7(33) 11.060.7(36) 11.060.8(42) z-score/p LL 1 21.97/0.02 21.27/0.10 0.56/0.28 - 20.88/0.19 z-score/p MLDG 1679 - - - 1.27/0.10 0.74/0.23 ERECT{ 7.160.5(11) 9.160.7(11) 10.060.6(16) 12.860.9(18) 11.561.1(14) z-score/p LL 1 20.77/0.23 21.05/0.08 20.97/0.17 - 20.88/0.19 z-score/p MLDG 1679 - - - 20.97/0.17 20.09/0.46

"Sample abbreviations and compositions see Table 4 and [9–10]; m6s(n); z-test results do not employ Bonferroni correction as per [56]. {Comparative samples from Shang et al. [9]. {Data compiled by the authors from literature (see Table 4 and Text S2). 1Mean of left and right used in z-test. doi:10.1371/journal.pone.0031918.t018

then with successively stronger solutions until the solution is alteration are outlined in [63–65]. Bulk sediment samples were clear or until all humic acids are removed, taken from every single excavated stratigraphic unit during N 2 M HCl at room temperature for 4 hours to remove any excavation. These bulk samples were divided into sub-samples, air dried and sieved to remove any large non-magnetic particles (i.e. atmospheric CO2, which was absorbed by the samples during the alkali treatment. clasts). The sieved bulk sub-samples were then subjected to a range of mineral magnetic measurements. Low temperature The cleaned charcoal pieces are finally placed into an oven at and room temperature dual frequency magnetic susceptibility 60uC for 2–3 days to dry and then taken for combustion using measurements were undertaken on a Bartington MS2 system. routine methods for conversion of charcoal to graphite [60]. A Isothermal remanent magnetisation (IRM) acquisition and backfield portion of each graphite sample was used to determine d13C for mass curves, hysteresis loops and thermomagnetic curves were run on a fractionation correction from the graphitisation process. Measured Magnetic Measurements Variable field Translation Balance (MM- AMS 14C/13C ratios are converted to conventional radiocarbon VFTB). ages after background subtraction and d13C fractional correction. Radiocarbon ages (see Table 1) are given with 1 standard deviation Endocast rendering and volume estimation (1s) precisions ranging from 60.3 to 0.5. All radiocarbon ages were A virtual endocast of MLDG 1704 was generated from converted to calibrated calendar ages BP (before-present, 1950) computed tomography (CT) data in Mimics (Ver. 13.02) by: using the CALIB 6.02 calibration software and the IntCal09 data sets [62]. All calendar age errors quoted in this paper are given as 2 1. Segmenting out extraneous material and generating a mask for standard deviation errors (2s). Table 1 provides radiocarbon ages MLDG 1704, and calibrated calendar ages for each charcoal sample measured by 2. Generating a cutting plane and converting this mask into a 3D AMA. Table S1 provides ancillary information pertaining to sample object, pretreatment, graphite AMS mass and d13Cvaluesusedtocorrect 3. Positioning the 3D object such that it closed the open region of AMS radiocarbon data from Maludong. the cranium, 4. Generating a mask from the repositioned 3D of the cutting Archaeomagnetics plane, Detailed theories and methods related to the use of magnetic 5. Combining the mask of MLDG 1704 with that of the cutting measurements for reconstructing palaeoclimate and anthropogenic plane,

PLoS ONE | www.plosone.org 25 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

Table 19. Comparison of maxillary dental crown brain. Four of the polygons were defined by 100 landmarks (9 Type buccolingual diameters (mm) (significant z-scores in bold)." II, 91 slid semilandmarks), with the remaining 8 polygons defined by 25 landmarks (9 Type II, 16 slid semilandmarks). Once the landmarks were placed on all of the crania, Template Optimisation Fossil/Sample P4 M1 M3 was used to create the ‘mean’ endocranial whole surface shape of these six humans (Figure S3). Template Optimisation has been LL 1 11.0 11.7 - shown to be accurate in reproducing the target mesh shapes [68]. MLDG 1747 - - 12.5 The mean endocranial shape was registered with NMB 1204 EAEHS{ 10.5(4) 12.360.7(8) 11.761.3(5) using an Iterative Closest Point (ICP) registration algorithm [69– z-score/p LL 1 20.49/0.31 20.81/0.22 - 71], to place it in a 3D space relevant to that of the other human endocrania. The modern human endocrania and that of MLDG z-score/p MLDG 1747 - - 0.56/0.30 1704 were ICP registered with the mean endocranium to minimise { Early Upper Palaeolithic --11.661.1(27) any orientation differences between endocranial specimens and the z-score/p MLDG 1747 - - 0.80/0.21 mean shape (Figure S4). STLs of the registered ‘mean’ the San and Middle Palaeolithic HS{ - - 11.9(1) MLDG 1704 endocasts were imported into Mimics and a cutting Neandertals{ 10.060.8(21) 12.060.8(34) 11.961.4(29) plane was generated and positioned as above, with the endocast of MLDG 1704 superimposed (Figure S4). This was used to separate z-score/p LL 1 1.40/0.08 20.37/0.35 - that part of the mean San endocast that was not preserved in z-score/p MLDG 1747 - - 0.42/0.33 MLDG 1704. The volume of this separated portion amounted to Qafzeh-Skhul{ 10.260.8(10) 12.260.7(18) 11.760.6(6) 39% of the original total brain volume for the mean San endocast. z-score/p LL 1 0.95/0.18 20.70/0.24 - The total brain volume of MLDG 1704 is estimated to be 1327 cm3 z-score/p MLDG 1747 - - 0.42/0.33 assuming similarity in proportions between the two. Early Upper Palaeolithic HS{ 9.960.6(25) 12.260.8(37) 11.661.1(27) Supporting Information z-score/p LL 1 1.80/0.04 20.62/0.27 - z-score/p MLDG 1747 - - 0.80/0.21 Figure S1 Endocasts generated from computed tomog- ERECT{ 11.561.0(16) 13.061.0(11) 11.860.9(8) raphy. a, MLDG 1704, and San crania: b, NMB 4. c, NMB 1271. d, NMB 1640. e, NMB 1204. f, NMB 1707. g, NMB 1240. z-score/p LL 1 20.49/0.31 21.24/0.12 - (TIF) z-score/p MLDG 1747 - - 0.73/0.24 Figure S2 Varying views of the BW 1204 modern human "Sample abbreviations and compositions see Table 4 and [58–59]; m6s(n); z- (San) endocranium specimen showing the landmark and test results do not employ Bonferroni correction as per [56]. slid semilandmark template as applied to each of the 6 {Data compiled by the authors from literature (see Table 4 and Text S2). {Comparative sample statistics from Trinkaus et al. [58–59]. modern human specimens. Aspect viewed: A) inferior, B) doi:10.1371/journal.pone.0031918.t019 superior, C) frontal, D) L frontal-inferior E) lateral F) L inferior- lateral. 6. Using the ‘cavity fill’ tool to create a partial endocast from this (TIF) combined mask, Figure S3 Landmark template. A) applied to BW 1204 (as in 7. A 3D surface mesh was then generated from this mask of the Figure S2), B) same landmark template applied to specimen BW endocast and imported into Strand7 (ver. 2.4), and 1240, C) and mean modern (San) endocranial shape generated 8. A solid mesh of the partial endocast was then created in Strand7 from the average landmark configuration of the 6 modern human and the volume taken from the model summary. specimens. (TIF) Six endocasts and their respective volumes were generated from CT scans of complete Holocene age southern African San crania Figure S4 Registered mean San (dark gray) and MLDG using this same general approach (Figure S1). In these instances ‘holes’ 1704 (light gray) endocasts superimposed showing in the masks of the crania representing nerves and blood vessels were cutting plane. filled before applying the ‘cavity fill’ tool to produce the endocasts. (TIF) A template of Type I, Type II and Type III [66] landmark Table S1 Radiocarbon data for Maludong. points was created to capture the whole surface morphology of the (DOCX) six modern human endocasts (Figure S2). Warping of cranial exterior surface morphology using a mixture of landmark points Text S1 Archaeomagnetics - results. and slid semi-landmarks has been shown to be highly effective at (DOCX) reproducing target cranial shape [67]. Here we apply a similar Text S2 Additional references: sources of metrical and methodology, utilising landmarks, pseudo-landmarks and slid morphological data and dating estimates. semi-landmarks, to these endocrania. The landmark template (DOCX) was designed to capture as much as possible of the endocranial shape that was common to all six modern humans. Acknowledgments Our landmark template consisted of 715 landmarks. We used 33 single points (Type I and II landmarks), 9 curves (the beginning and We are very grateful to Ruliang Pan, Yang Decong (Yunnan Institute for end of the curves were defined by Type II landmarks, with 8 Cultural Relics and Archaeology), He Guangshu (Vice Director, Education, Science, Culture and Health Committee of CPPC, Yunnan additional Type III semilandmarks slid between these across the Province) and Fan Mian (former Director, Honghe Institute of Cultural endocranial surface) and 12 polygon regions (9 user defined Type II Relics) for assistance with establishing and managing the project. We thank landmarks with additional slid semilandmarks). The polygon regions You Pingping for research assistance during the duration of the project, were used to capture the morphology of the different lobes of the Zhang Tao (Yunnan University) for assistance with fieldwork at Maludong

PLoS ONE | www.plosone.org 26 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China in 2008, and Deng Yamin (First Hospital Affiliated with Kunming Medical Author Contributions College) for undertaking CT-scans of human fossils. The staff of the University of Liverpool Geomagnetism Laboratory, particularly Mimi Hill Conceived and designed the experiments: DC JX AH BK PT BZ DF ZY and John Shaw, are thanked for assistance with undertaking some of the JH LY GC SB SW HS WP HS NR. Performed the experiments: DC JX archaeomagnetic analyses. We thank James Brink for permission to use AH BK PT BZ DF ZY JH LY GC SB SW HS WP HS NR. Analyzed the CT-Scans of Holocene San crania. Finally, we wish to thank two data: DC JX AH BK PT BZ DF ZY JH LY GC SB SW HS WP HS NR. anonymous reviewers whose comments improved our work. Wrote the paper: DC JX AH BK PT BZ DF ZY JH LY GC SB SW HS WP HS NR.

References 1. Trinkaus E (2005) Early modern humans. Annu Rev Anthropol 34: 207–230. 28. Trinkaus E (2007) European early modern humans and the fate of the 2. Wu X, Poirier FE (1995) Human Evolution in China. New York: Oxford Univ. Neandertals. Proc Natl Acad Sci USA 104: 7367–7372. Press. 29. Howells WW (1996) Howell’s craniometric data on the internet. Am J Phys 3. Kaifu Y, Fujita M (2012) Fossil record of early modern humans in East Asia. Anthropol 101: 441–442. Quat Int 248: 2–11. 30. Dobson SD, Trinkaus E (2002) Cross-sectional geometry and morphology of the 4. Bra¨uer G, Mı´misson K (2004) Morphological affinities of early modern crania mandibular symphysis in Middle and Late Pleistocene Homo. J Hum Evol 43: from China. In: Zona Arqueolo´gica, Miscela´nea en homenaje a Emiliano 67–87. Aguirre, Vol. 3, Nu´mero 4, Paleoanthopologı´a, Museuo Arqueolo´gico Regional, 31. Schwartz JH, Tattersall I (2000) The human chin revisited: what is it and who Alcala´de Henares. pp 59–69. has it? J Hum Evol 38: 367–409. 5. Barker G, Barton H, Bird M, Daly P, Datan I, et al. (2007) The ‘human 32. Crevecoeur I, Trinkaus E (2004) From the Nile to the Danube: a comparison of revolution’ in lowland tropical Southeast Asia: the antiquity and behavior of the Nazlet Khater 2 and Oase 1 early modern human mandibles. Anthropologie anatomically modern humans at Niah Cave (Sarawak, Borneo). J Hum Evol 52: 152: 203–213. 243–261. 33. Harvati K, Panagopoulou E, Karkanas P (2003) First remains from 6. Dizon E, De´troit F, Se´mah F, Falgue`res C, Hameau S, et al. (2002) Notes on the : the evidence from Lakonis. J Hum Evol 45: 465–473. morphology and age of the Tabon Cave fossil Homo sapiens. Curr Anthropol 43: 34. Bu¨rklein S, Breuer D, Scha¨fer E (2011) Prevalence of taurodont and pyramidal 660–666. molars in a German population. J Endodont 37: 158–162. 7. De`troit F, Dizon E, Falgue´res C, Hameau S, Ronquillo W, et al. (2004) Upper 35. Kupczik K, Hublin JJ (2010) Mandibular molar root morphology in Pleistocene Homo sapiens from the Tabon cave (Palawan, The Philippines): and late Pleistocene and recent Homo sapiens. J Hum Evol 59: description and dating of new discoveries. C R Palevol 3: 705–712. 525–541. 8. Mijares AM, De`troit F, Piper P, Gru¨n R, Bellwood P, et al. (2010) New evidence 36. Bailey SE (2002) A closer look at Neanderthal postcanine dental morphology: the mandibular dentition. Anat Rec 269: 148–156. for a 67,000-year-old human presence at Callao Cave, Luzon, Philippines. nd J Hum Evol 59: 123–132. 37. Enlow DH, Hans MG (2008) Essentials of facial growth. 2 Ed. Needham, Ann 9. Shang H, Tong H, Zhang S, Chen F, Trinkaus E (2007) An early modern Arbor, Mich, USA. human from , Zhoukoudian, China. Proc Natl Acad Sci 104: 38. Sjøvold T (1984) A report on the heritability of some cranial measurements and 6573–6578. non-metric traits. In: van Vark GN, Howells WW, eds. Multivariate Statistical 10. Liu W, Jin C-Z, Zhang Y-Q, Cai YJ, Xing S, et al. (2010) Human remains from Methods in Physical Anthropology, D. Reidel Publishing, Dordrecht, the Zhirendong, South China, and modern human emergence in East Asia. Proc Netherlands. pp 223–246. Natl Acad Sci 107: 19201–19206. 39. Carson AE (2006) Maximum likelihood estimation of human craniometric 11. Dennell R (2010) Early Homo sapiens in China. Nature 468: 512–513. heritabilities. Am J Phys Anthropol 131: 169–180. 12. Zhong H, Shi H, Qi X-B, Duan Z-Y, Tan P-P, et al. (2011) Extended Y 40. Sherwood RJ, Duren DL, Demerath EW, Czerwkinski SA, Siervogel RM, et al. chromosome investigation suggests postglacial migrations of modern humans (2008) Quantitative genetics of modern human cranial variation. J Hum Evol 54: into East Asia via the northern route. Mol Biol Evol 28: 717–727. 909–914. 41. Townsend GC, Brown T (1978) Heritability of permanent tooth size. Am J Phys 13. Kong Q-P, Sun C, Wang H-W, Zhao M, Wang WZ, et al. (2011) Large-scale Anthropol 49: 497–504. mtDNA screening reveals a surprising matrilineal complexity in East Asia and its 42. Weaver TD, Roseman CC, Stringer CB (2007) Were neandertal and modern implications to the peopling of the region. Mol Biol Evol 28: 513–522. human cranial differences produced by natural selection or genetic drift? J Hum 14. Stoneking M, Delfin F (2010) The human genetic history of East Asia: weaving a Evol 53: 135–145. complex tapestry. Curr Biol 20: R188–R193. 43. Smith FH (1992) Models and realities in modern human origins: the African 15. Rasmussen M, Guo X, Wang Y, Lohmueller KE, Rasmussen S, et al. (2011) An fossil evidence. Phil Trans Biol Sci 337: 243–250. Aboriginal Australian genome reveals separate human dispersals into Asia. 44. Lam YM, Pearson OM, Smith CM (1998) Chin morphology and sexual Science 334: 94–98. dimorphism in the fossil hominid mandible sample from Klasies River Mouth. 16. Reich D, Green RE, Kircher M, Krause J, Patterson N, et al. (2010) Genetic Am J Phys Anthropol 100: 545–557. history of an archaic hominin group from Denisova Cave in Siberia. Nature 468: 45. Crevecoeur I, Rougier H, Grine F, Froment A (2009) Modern human cranial 1053–1060. diversity in the late Pleistocene of Africa and Eurasia: evidence from Nazlet 17. Reich D, Patterson N, Kircher M, Delfin F, Nandineni MR, et al. (2011) Khater, Pestera cu Oase, and Hofmeyr. Am J Phys Anthropol 140: 347–358. Denisova Admixture and the First Modern Human Dispersals into Southeast 46. Harvati K, Stringer C, Gru¨n R, Aubert M, Allsworth-Jones P, et al. (2011) The Asia and Oceania. Am J Hum Genet 89: 516–528. Later Stone Age calvaria from Iwo Eleru, Nigeria: morphology and chronology. 18. Abi-Rached L, Jobin MJ, Kulkarni S, McWhinnie A, Dalva K, et al. (2011) The PLoS One 6: e24024. shaping of modern human immune systems by multiregional admixture with 47. Jacob Z, Meyer MC, Roberts RG, Aldeias V, Dibble H, et al. (2011) Single- archaic humans. Science 334: 89–94. grain OSL dating at La Grotte des Contrebandiers (‘Smugglers’ Cave’), 19. Zhang X, Liang Z, Liechang Y, Zhende B (1991) Human fossils and the Morocco: improved age constraints for the levels. J Archaeol palaeoculture from Mengzi. In: Materials of Human Origin and Prehistoric Sci 38: 3631–3643. Culture from Yunnan, Kunming, Yunnan Provincial Museum, Yunnan 48. Millard AR (2008) A critique of the chronometric evidence for hominid fossils: I. Renming Press. pp 234–246. Africa and the Near East 500-50 ka. J Hum Evol 54: 848–874. 20. Clark PU, Dyke AS, Shakun JD, Carlson AE, Clark J, et al. (2007) The Last 49. Grine FE, Bailey RM, Harvati K, Nathan RP, Morris AG, et al. (2007) Late Glacial Maximum. Science 325: 710–714. Pleistocene human skull from Hofmeyr, South Africa, and modern human 21. Carbonell E, Bermu´dez de Castro JM, Arsuaga JL, Allue E, Bastir M, et al. origins. Science 315: 226–229. (2005) An Early Pleistocene hominin mandible from Atapuerca-TD6, . 50. Coppens Y, Tsveendorj D, Demeter F, Turbat T, Giscard P-H (2008) Discovery Proc Natl Acad Sci USA 102: 5674–5678. of an archaic Homo sapiens skullcap in Northeast Mongolia. C R Palevol 7: 51–60. 22. Lahr MM (1996) The evolution of modern human diversity: a study of cranial 51. Mava´rez J, Linares M (2008) Homoploid hybrid speciation in animals. Mol Ecol variation. Cambridge: Cambridge University Press. 17: 4181–4185. 23. Tattersall I, Schwartz J (2008) The morphological distinctiveness of Homo sapiens 52. Gunz P, Bookstein FL, Mitteroecker P, Stadlmayr A, Horst S, et al. (2009) Early and its recognition in the fossil record: clarifying the problem. Evol Anthropol modern human diversity suggests subdivided population structure and a 17: 49–54. complex out-of-Africa scenario. Proc Natl Acad Sci USA 106: 6094–6098. 24. Smith FH, Ranyard GC (1980) Evolution of the supraorbital region in upper 53. Hellstrom JC (2003) Rapid and accurate U/Th dating using parallel ion- Pleistocene fossil hominids from south-central Europe. Am J Phys Anthropol 53: counting multi-collector ICP-MS. J Anal At Spectrom 18: 1346–1351. 589–610. 54. Cheng H, Edwards RL, Gallup CD, Richards DA, Asmerom Y (2000) The half- 25. Sokal RR, Rohlf FJ (1981) Biometry, 2nd ed. New York: W. H. Freeman. lives of uranium-234 and thorium-230. Chem Geol 169: 17–33. 26. Grine FE, Pearson OM, Klein RG, Rightmire GP (1998) Additional human 55. Hellstrom JC (2006) U-Th dating of with high initial Th-230 using fossils from Klasies River Mouth, South Africa. J Hum Evol 35: 97–107. stratigraphical constraint. Quat Geochron 1: 289–295. 27. Darroch JN, Mosimann JE (1985) Canonical and principal components of 56. Nakagawa S (2004) A farewell to Bonferroni: the problems of low statistical shape. Biometrika 72: 241–252. power and publication bias. Behav Ecol 15: 1044–1045.

PLoS ONE | www.plosone.org 27 March 2012 | Volume 7 | Issue 3 | e31918 Human Remains from Southwest China

57. Howells WW (1989) Skull Shapes and the Map. Homo. Papers of the Peabody 65. Herries AIR, Fisher E (2010) Multi-dimensional modelling of magnetic Museum of Archaeology and Ethnology, vol. 79, Cambridge, Mass.: Peabody mineralogy as a proxy for fire use and spatial patterning: evidence from the Museum. Middle Stone Age sea cave site of 13B (Western Cape, South 58. Trinkaus E, Milota S, Rodrigo R, Mircea G, Moldovand O (2003) Early Africa). J Hum Evol 59: 306–320. modern human cranial remains from the Pestera cu Oase, Romania. J Hum 66. Bookstein FL (1991) Morphometric Tools for Landmark Data: Geometry and Evol 45: 245–253. Biology. Cambridge: Cambridge University Press. 59. Trinkaus E, Biglari F, Mashkour M, Monchot H, Reyss J-J, et al. (2008) Late 67. Gunz P, Mitteroecker P, Neubauer S, Weber GW, Bookstein FL (2009) Pleistocene human remains from Cave, western . Am J Phys Principles for the virtual reconstruction of hominin crania. J Hum Evol 57: Anthropol 135: 371–378. 48–62. 60. Fink D, Hotchkis M, Hua Q, Jacobsen G, Smith AM, et al. (2004) The 68. Ruto A (2009) Dynamic Human Body Modeling and Animation. Unpublished PhD ANTARES AMS facility at ANSTO. In Nuclear Instruments and Methods in Thesis, University of London. Physics Research Section B: Beam Interactions with Materials and Atoms, 223– 69. Besl PJ, Mckay ND (1992) A method for registration of 3-D shapes. IEEE Trans 224. pp 109–115. Pat Anal Mach Intel 14: 239–256. 61. Hua Q, Jacobsen GE, Zoppi U, Lawson EM, Williams AA, et al. (2001) Progress 70. Parr WCH (2009) Evolutionary and Functional Anatomy of the Hominoid Astragalus – in radiocarbon target preparation at the ANTARES AMS Centre. Radiocarbon New Approaches Using Laser Scanning Technologies and 3D Analyses. Unpublished PhD 43: 275–282. Thesis, University of London. 62. Stuiver M, Reimer PJ, Reimer R (2005) CALIB 6.0. WWW program and 71. Parr WCH, Ruto A, Soligo C, Chatterjee HJ (2011) Allometric shape vector documentation. Available: http://radiocarbon.pa.qub.ac.uk/calib/. projection: A new method for the identification of allometric shape characters 63. Walden J, Oldfield F, Smith J (1999) Environmental Magnetism: a Practical and trajectories applied to the human astragalus (talus). J Theor Biol 272: 64–71. Guide. Quaternary Research Association Technical Guide 6. 64. Herries AIR (2009) New approaches for integrating palaeomagnetic and mineral magnetic methods to answer archaeological and geological questions on Stone Age sites. In: Fairbrain A, O’Connor S, Marwick B, eds. Terra Australis 28. Canberra, Australia: The Australian National University Press.

PLoS ONE | www.plosone.org 28 March 2012 | Volume 7 | Issue 3 | e31918

Minerva Access is the Institutional Repository of The University of Melbourne

Author/s: Curnoe, D; Xueping, J; Herries, AIR; Kanning, B; Tacon, PSC; Zhende, B; Fink, D; Yunsheng, Z; Hellstrom, J; Yun, L; Cassis, G; Bing, S; Wroe, S; Shi, H; Parr, WCH; Shengmin, H; Rogers, N

Title: Human Remains from the Pleistocene-Holocene Transition of Southwest China Suggest a Complex Evolutionary History for East Asians

Date: 2012-03-14

Citation: Curnoe, D., Xueping, J., Herries, A. I. R., Kanning, B., Tacon, P. S. C., Zhende, B., Fink, D., Yunsheng, Z., Hellstrom, J., Yun, L., Cassis, G., Bing, S., Wroe, S., Shi, H., Parr, W. C. H., Shengmin, H. & Rogers, N. (2012). Human Remains from the Pleistocene-Holocene Transition of Southwest China Suggest a Complex Evolutionary History for East Asians. PLOS ONE, 7 (3), https://doi.org/10.1371/journal.pone.0031918.

Persistent Link: http://hdl.handle.net/11343/264875

File Description: Published version License: CC BY